cme article journal watch paediatrics...
TRANSCRIPT
ISSN 1015-4361(PHILIPPINES)
PhilippinesJAN/FEB 2012 Vol. 38 No. 1
JOURNAL OF PAEDIATRICS, OBSTETRICS & GYNAECOLOGY
Your partner in paediatric and O&G practice
JUL/AUG 2013 Vol. 39 No. 4 Your partner in paediatric and O&G practice
www.jpog.com
CME ARTICLE
Hormonal Contraception and Cancers
JOURNAL WATCH
PAEDIATRICS
Management of Hearing Loss in Children
Constipation in Infants and Children
GYNAECOLOGY
Ovarian Cancer: Current Management and Future Directions
OBSTETRICS
Management of Early Pregnancy
Complications
JPOG_JulAug13_CVR_FINAL.indd 1 8/5/13 2:00 PM
JUL/AUG 2013
Vol. 39 No. 4
JOURNAL OF PAEDIATRICS, OBSTETRICS & GYNAECOLOGY
Journal Watch
133 • Calcium intake and mortality in Swedish women
• Outcome of extreme preterm birth in England, 1995–2006
134 • New anti-interleukin therapies for systemic JIA
• CPAP vs surfactant and higher vs lower oxygen saturation for extremely preterm infants: Outcomes at 18–22 months
135 • Paromomycin for cutaneous leishmaniasis
• HMPV infection in young children in the US
136 • Risk factors for stillbirth • Fetal macrosomia in developing countries
Board Director, Paediatrics
Professor Pik-To CheungAssociate ProfessorDepartment of Paediatrics and Adolescent MedicineThe University of Hong Kong
Board Director, Obstetrics and Gynaecology
Professor Pak-Chung HoHead, Department of Obstetrics and GynaecologyThe University of Hong Kong
Editorial Board Professor Biran AffandiUniversity of Indonesia
Dr Karen Kar-Loen ChanThe University of Hong Kong
Professor Oh Moh ChayKK Women’s and Children’s Hospital, Singapore
Associate Professor Anette JacobsenKK Women’s and Children’s Hospital, Singapore
Professor Rahman JamalUniversiti Kebangsaan Malaysia
Dato’ Dr Ravindran JegasothyHospital Kuala Lumpur, Malaysia
Associate Professor Kenneth KwekKK Women’s and Children’s Hospital, Singapore
Dr Siu-Keung LamPrestige Medical Centre, Hong Kong
Professor Terence LaoChinese University of Hong Kong
Dr Kwok-Yin LeungThe University of Hong Kong
Dr Tak-Yeung LeungChinese University of Hong Kong
Professor Tzou-Yien LinChang Gung University, Taiwan
Professor Somsak LolekhaRamathibodi Hospital, Thailand
Professor Lucy Chai-See LumUniversity of Malaya, Malaysia
Professor SC NgNational University of Singapore
Professor Hextan Yuen-Sheung NganThe University of Hong Kong
Professor Carmencita D PadillaUniversity of the Philippines Manila
Professor Seng-Hock QuakNational University of Singapore
Dr Tatang Kustiman SamsiUniversity of Tarumanagara, Indonesia
Professor Alex SiaKK Women’s and Children’s Hospital, Singapore
Dr Raman SubramaniamFetal Medicine and Gynaecology Centre, Malaysia
Professor Walfrido W Sumpaico MCU-FDT Medical Foundation, Philippines
Professor Cheng Lim TanKK Women’s and Children’s Hospital, Singapore
Professor Kok Hian TanKK Women’s and Children’s Hospital, Singapore
Professor Surasak TaneepanichskulChulalongkorn University, Thailand
Professor Eng-Hseon TayThomson Women Cancer Centre, Singapore
Professor PC WongNational University of Singapore
Adjunct Professor George SH YeoKK Women’s and Children’s Hospital, Singapore
Professor Hui-Kim YapNational University of Singapore
Professor Tsu-Fuh YehChina Medical University, Taiwan
133
136
JPOG JUL/AUG 2013 • i
JPOG_JulAug_2013_TOC_Final.indd 1 8/5/13 1:52 PM
JUL/AUG 2013
Vol. 39 No. 4
JOURNAL OF PAEDIATRICS, OBSTETRICS & GYNAECOLOGY
Enquiries and Correspondence
137
JPOG JUL/AUG 2013 • ii
Publisher Ben YeoPublication Manager Marisa LamManaging EditorGreg TownAssociate Editor Grace LingDesignersAgnes Chieng, Sam ShumProduction Edwin Yu, Ho Wai Hung, Steven Cheung Circulation Christine ChokAccounting Manager Minty KwanAdvertising CoordinatorRachael Tan
Published by: MIMS Asia Pacific27th Floor, OTB Building160 Gloucester Road, Wan Chai, Hong KongTel: (852) 2559 5888 Email: [email protected]
PUBLISHER: Journal of Paediatrics, Obstetric & Gynaecology (JPOG) is published 6 times a year by MIMS Pte Ltd. CIRCULATION: JPOG is a controlled circulation for medical practitioners in South East Asia. It is also available on subscription to members of allied professions. SUBSCRIPTION: The price per annum is US$42 (surface mail, students US$21) and US$48 (overseas airmail, students US$24); back issues US$8 per copy. EDITORIAL MATTER published herein has been prepared by professional editorial staff. Views expressed are not necessarily those of MIMS Pte Ltd. Although great care has been taken in compiling and checking the information given in this publication to ensure that it is accurate, the authors, the publisher and their servants or agents shall not be responsible or in any way liable for the continued currency of the information or for any errors, omissions or inaccuracies in this publication whether arising from negligence or otherwise howsoever, or for any consequences arising therefrom. The inclusion or exclusion of any product does not mean that the publisher advocates or rejects its use either generally or in any particular field or fields. COPYRIGHT: © 2013 MIMS Pte Ltd. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, in any language, without written consent of copyright owner. Permission to reprint must be obtained from the publisher. ADVERTISEMENTS are subject to editorial acceptance and have no influence on editorial content or presentation. MIMS Pte Ltd does not guarantee, directly or indirectly, the quality or efficacy of any product or service described in the advertisements or other material which is commercial in nature. Philippine edition: Entered as second-class mail at the Makati Central Post Office under Permit No. PS-326-01 NCR, dated 9 Feb 2001. Printed by Fortune Printing International Ltd, 3rd Floor, Chung On Industrial Building, 28 Lee Chung Street, Chai Wan, Hong Kong.
ChinaYang XuanTel: (86 21) 6157 3888Email: [email protected]
Hong KongKristina Lo-Kurtz, Jacqueline Cheung, Marisa Lam, Miranda WongTel: (852) 2559 5888Email: [email protected]
IndiaMonica BhatiaTel: (91 80) 2349 4644 Email: [email protected]
KoreaChoe Eun YoungTel: (82 2) 3019 9350Email: [email protected]
IndonesiaHafta Hasibuan, Sri Damayanti, Ritta PamolangoTel: (62 21) 729 2662Email: [email protected]
MalaysiaMeera Jassal, Lee Pek Lian, Sheena Subash, Grace YeohTel: (60 3) 7954 2910Email: [email protected]
PhilippinesMarian Chua, Kims Pagsuyuin, Rowena Belgica, Philip KatipunanTel: (63 2) 886 0333Email: [email protected]
SingaporeJason Bernstein, Carrie Ong, Josephine Cheong, Melanie NyamTel: (65) 6290 7400Email: [email protected]
ThailandWipa SriwijitchokTel: (66 2) 741 5354Email: [email protected]
VietnamNguyen Thi Lan Huong, Nguyen Thi My DungTel: (84 8) 3829 7923Email: [email protected]
Europe/USAKristina Lo-KurtzTel: (852) 2116 4352Email: [email protected]
Review ArticlePaediatrics
137 The Management of Hearing Loss in Children
Universal neonatal hearing screening aims to detect the 1 in 1,000 babies born in the UK with a permanent hearing loss detectable at birth. However, children may present later to the paediatrician with hearing difficulties. This article aims to discuss the clinical assessment of hearing and provides an overview of the management options available in the treatment of hearing loss.
Marianne D Elloy, Andrew H Marshall
Review ArticleObstetrics
146 Management of Early Pregnancy Complications
Complications of early pregnancy are common, including pregnancy loss, threatened miscarriage, ectopic pregnancy, molar pregnancy and hyperemesis. This review discusses the different presentations, diagnoses and management of the common problems complicating early pregnancy.
Harriet Pugsley, Judith Moore
146
JPOG_JulAug_2013_TOC_Final.indd 2 8/5/13 1:52 PM
JOURNAL OF PAEDIATRICS, OBSTETRICS & GYNAECOLOGY
JPOG JUL/AUG 2013 • iii
Review ArticlesComprehensive reviews providing the latest clinical information on all aspects of the management of medical conditions affecting children and women.
Case StudiesInteresting cases seen in general practice and their management.
Pictorial MedicineVignettes of illustrated cases with clinical photographs.
For more information, please refer to the Instructions for Authors on our website www.jpog.com, or contact:The EditorMIMS Pte Ltd, 6 Shenton Way, #15-08 Tower 2, Singapore 068809Tel: (65) 6290 7400 Fax: (65) 6290 7401 E-mail: [email protected]
Lisa Low, Illustrator
JUL/AUG 2013
Vol. 39 No. 4
The Cover:Early Pregnancy Complications
© 2013 MIMS Pte Ltd
Review ArticleGynaecology
155 Ovarian Cancer: Current Management and Future Directions
Ovarian cancer has the highest mortality of all the gynaecological malignancies. Treatment of advanced epithelial ovarian cancer usually involves debulking surgery and chemotherapy. Treatment may prolong life and palliate symptoms but it is rarely curative. New treatments are constantly being developed and offer the hope of improved outcomes.
Siân E Taylor, John M Kirwan
Review ArticlePaediatrics
164 Constipation in Infants and Children
Constipation is a common problem in children and is usually functional, related to stool-withholding. Successful management requires parent education, behavioural strategies, laxative agents (often long term) and ongoing review.
Taya Dowling, Scott Nightingale
Continuing Medical Education
169 Hormonal Contraception and Cancers
This article reviews the evidence regarding the relationship between hormonal contraceptive use and the development of cancer, with the discussion focusing mainly on carcinoma of the breast and female genital tract.
Wong Yuen Kwan Alice
155
169
Philippines
JPOG_JulAug_2013_TOC_Final.indd 7 8/5/13 1:52 PM
Continuing Medical Education
JPOG JUL/AUG 2013 • 169
Hormonal Contraception and Cancers Wong Yuen Kwan Alice, MBBS, FRCOG, FHKAM(O&G), FHKCOG, Cert HKCOG(Reprod Med)
Oral contraceptive use puts women at risk for certain carcinomas.
INTRODUCTION
The use of hormones has provided great
convenience to a woman’s life, be it for
therapeutic use or as a lifestyle drug, ie,
contraceptives. However, the duration of
hormone use is frequently long, in terms
of years. Concerns have been raised about
the possibility of a relationship between
cancer development and long-term hormo-
nal influence. This article reviews the evi-
dence regarding the relationship between
hormonal contraceptive use and the devel-
opment of cancer, with the discussion fo-
cusing mainly on carcinoma of the breast
and female genital tract.
HISTORICAL EVIDENCE
The first report dated back to 1972 when
combined oral contraceptive pills (COC)
containing mestranol and norethynodrel
appeared to cause a case of metastatic
breast cancer in a female rhesus monkey.1
Soon after, there were further similar re-
ports of development of breast cancer in
beagles and rodents after exposure to hor-
mones contained in today’s COC.2–4
In June 2005, the International Agen-
cy for Research on Cancer (IARC) Working
Group of the World Health Organization
(WHO) met in Lyon, France, and classified
combined oral contraceptives and com-
bined oestrogen-progestogen hormone
therapy as ‘carcinogenic’ to humans.
THEORY OF ‘CARCINOGENESIS’
Carcinogenesis involves two steps, name-
ly, initiation and promotion. Most of the
studies on the relationship between hor-
mones and cancer development involved
the latter step. In 1989, Anderson et al
reported that nulliparous women who
took COC had a significantly higher rate of
breast cell division.5 It was also found that
COC caused a rise in epithelial cell pro-
liferation of the glandular breast,6 leading
to an increase in accumulation of random
genetic errors.7
However, whether these proliferat-
ing effects on normal epithelia, as a result
of replication error, may cause malignant
transformation has not yet been proven,
although DNA repair is hampered by ac-
tivated proliferation.8 With the end point
being ‘chromosomal mutation’, numer-
ous chromosomal aberrations have been
JPOG_JulAug_2013_CME_PH_Final_Hormonal Contraception and Cancers.indd 169 8/5/13 2:13 PM
JPOG JUL/AUG 2013 • 170
observed with natural and synthetic oes-
trogens and progestogens. There was no
proof of sufficient strength to show that
these proliferative effects could induce
tumours. Hormonal tumour promotion also
cannot be discriminated from a causal re-
lation with breast tumour induction.
Moreover, toxicity studies used
animal models. Species-specific effects
might not allow for extrapolation of the ef-
fects to humans. Supraphysiological doses
had been used in animal studies and this
might not reflect actual clinical use. Pos-
sibility of genetic predisposition and other
environmental factors were not taken into
account.
CARCINOMA OF THE BREAST
In 1981, Pike et al reported that women
who took COC for 4 years or more prior to
their first full-term pregnancy experienced
a 125% increased risk of developing carci-
noma of the breast and a 250% increase
in risk with 8 years or more of COC use.9
Similarly, in 1989, Chilvers et al reported
that women under the age of 36 who used
COC for at least 4 years before their first
full-term pregnancy had at least 44% in-
creased risk of breast cancer.10
However, the Cancer and Steroid Hor-
mone Study (CASH), which was one of the
largest case-control studies in the 1980s,
Combined oral contraceptives play a role in the promotion of carcinogenesis.
found no association between breast can-
cer and COC use for women up to the age
of 54. Risk was found only among a sub-
group of women who underwent menarche
before age 13 and used COC for more than
10 years before their first birth.11
In the 1990s, the Collaborative Group
on Hormonal Factors in Breast Cancer in
Oxford, UK, analysed individual data of
53,297 women with breast cancer and
100,239 women without breast cancer
from 54 studies conducted in 25 countries.
The results provided two strong conclu-
sions. First, while women are taking COC
and in a period of 10 years after stopping,
there is a small increase in the relative
risk (RR) of having breast cancer: RR of
1.24 (95% CI, 1.15–1.33) for current users;
RR of 1.16 (95% CI, 1.08–1.23) 1–4 years
after stopping; and RR of 1.07 (95% CI,
1.02–1.13) 5–9 years after stopping. Sec-
ond, there is no significant excess risk of
having breast cancer diagnosed 10 or more
years after stopping use (RR, 1.01 [95%
CI, 0.96–1.05]). The cancers diagnosed in
women who had used COC were less ad-
vanced clinically than those diagnosed in
never-users; the RR for tumours that had
spread beyond the breast compared with
localized tumours was 0.88 (95% CI, 0.81–
0.95). There was no pronounced variation
in the results for recency of use between
women with different background risks of
breast cancer, including women from dif-
ferent countries and ethnic groups, women
with different reproductive histories, and
those with or without a family history of
breast cancer. Other features of hormonal
use, such as duration of use, age at first
use, and the dose and type of hormone
JPOG_JulAug_2013_CME_PH_Final_Hormonal Contraception and Cancers.indd 170 8/5/13 2:13 PM
Continuing Medical Education
JPOG JUL/AUG 2013 • 171
within the contraceptives, had little ad-
ditional effect on breast cancer risk, once
recency of use had been taken into ac-
count.12
Other important studies worth men-
tioning include the Nurses’ Health Study
(1997),13 Women’s Lifestyle and Health
Cohort Study (2002),14 Oxford-Family Plan-
ning Association study (1981),15 Mayo
Clinic Meta-analysis (2006),16 and Royal
College of General Practitioners study
(2007).17 There were some inconsistencies
among their findings, and the increase in
RR shown by some of these studies was
modest. Possible confounders that may
be related to the use of high-dose COC
include the clinical practice during the
time of these studies and the possibil-
ity of the nature of recall bias. The study
by the Collaborative Group on Hormonal
Factors in Breast Cancer12 also shared a
similar problem. Another weakness of this
study was that it analysed pooled data
from studies which examined women with
breast cancer from as far back as the early
1970s. Taking data from studies which in-
terviewed women before the 1980s might
underestimate the risk of breast cancer
development because the latent period
for cancer development was too short and
few women had used COC for significant
periods of time prior to their first full-term
pregnancy in the late 1960s and early
1970s as compared with women of the
late 1970s and 1980s.
In 2005, the IARC classified COC and
combined oestrogen-progestogen hor-
mone therapy as carcinogenic to humans.
The Working Group mentioned a ‘slightly
increased risk of breast cancer in current
and recent users of hormonal contracep-
tives’. This risk disappears 10 years after
cessation of COC use and will be similar to
that in never-users.18 The Working Group
also acknowledged that their statement
does not meet the overall net public health
outcome, be this of a beneficial or ad-
verse effect other than cancer, and there
is no reason to change the current clini-
cal practice, particularly when the risks of
unwanted pregnancy are taken into con-
sideration.18
For BRCA mutation carriers, who al-
ready have a 50–80% increase in risk of
breast cancer, the use of COC will be of
concern. Among BRCA1 mutation carriers,
those who first used COC before 1975,
who used them before age 30, or who
used for 5 years or more might have an
increased risk of breast cancer. COC do not
appear to be associated with risk of breast
cancer in BRCA2 carriers; however, data to
support this are limited.19
On the use of depot medroxyproges-
terone acetate (MPA), pooled analysis of
two major case-control studies (one in
New Zealand20 and the other under the
auspices of the WHO21) found no increase
in risk for breast cancer. A currently unex-
plained pattern of increased risk in recent
users mimics that seen with COC.22
In a study involving completed ques-
Hormonal contraceptive use increases the relative risk of having breast cancer.
JPOG_JulAug_2013_CME_PH_Final_Hormonal Contraception and Cancers.indd 171 8/5/13 2:13 PM
JPOG JUL/AUG 2013 • 172
tionnaires from 17,360 levonorgestrel-
releasing intrauterine system (LNG-IUS)
users, there was no apparent associa-
tion between the length of time elapsed
from the LNG-IUS insertion up to 10 years
and yearly incidence of breast cancer in
the Finnish female population (data from
the Finnish Cancer Registry). A causal
relationship between LNG-IUS use and
occurrence of breast cancer was not sup-
ported.23
CARCINOMA OF THE UTERINE CORPUS
A meta-analysis of 10 case-control stud-
ies (published up to 1996; 1,728 cases and
6,243 controls) and another cohort study
(440,000 woman-years of observation)
both showed statistically reduced RR for
carcinoma of endometrium in COC users.
This RR was negatively associated with
the duration of COC use; the risk reduction
was 56% with 4 years’ use, 67% with 8
years’ use, and 72% with 12 years’ use.24
The Oxford-Family Planning Asso-
ciation (Oxford-FPA) contraceptive study,
which took place in 1968–2004, involved
540,000 woman-years of observation.
There were 50 women with carcinoma of
the uterine corpus in the control group and
27 women in the COC user group. The RR
for ever-users versus never-users was 0.3
(95% CI, 0.2–0.6). The risk was further
found to be negatively related to the du-
ration of COC use, ie, a RR of 0.6 (95%
CI, 0.3–1.1) for up to 48 months’ use, 0.4
(95% CI, 0.2–0.5) for 49–96 months’ use,
and 0.1 (95% CI, 0.0–0.4) for > 97 months’
use.25 The Royal College of General Prac-
titioners oral contraceptive study also
showed similar findings.17
Another meta-analysis of 11 epide-
miological studies found that the more
recent the use of COC, the lower the risk
for carcinoma of the uterine corpus. Such
protective effect from the previous use of
COC would attenuate with time after dis-
continuation. The RR was 0.33, 0.41 and
0.51 for 5, 10, and 20 years of ceasing, re-
spectively. Even after more than 20 years
of cessation of use, the protective effect
is still significant, ie, at 50% less than
non-users.24
Prolonged and unremitting mitotic
activity of the endometrium due to unop-
posed oestrogenic stimulation has been
proposed to be the cause of development
of the majority of cases of endometrial ad-
enocarcinoma. COC suppress endometrial
mitotic activity, leading to apoptosis, thus
reducing the risk of endometrial cancer.
The use of depot MPA is associated
with an 80% risk reduction of endometrial
adenocarcinoma, a level of protection even
greater than that observed with COC. The
effect was also found to be long-term.22
CARCINOMA OF THE OVARY
The Collaborative Group on Epidemiologi-
cal Studies of Ovarian Cancer published a
collaborative reanalysis of data from 45
epidemiological cohort and case-control
The protective effect of combined oral contraceptives against ovarian carcinoma is long-lasting.
JPOG_JulAug_2013_CME_PH_Final_Hormonal Contraception and Cancers.indd 172 8/5/13 2:13 PM
Continuing Medical Education
JPOG JUL/AUG 2013 • 173
studies, which included 23,257 women
with ovarian cancer and 87,303 controls
from 21 countries. It was found that 7,308
(31%) of the women with ovarian cancer
and 32,717 (37%) of the controls had used
COC, and the average duration of use was
4.4 and 5.0 years, respectively. The over-
all RR for ever-users versus never-users
was 0.73 (95% CI, 0.70–0.76). The longer
the duration of use, the lower the risk for
ovarian cancer development. The overall
RR decreased by 20% for each 5 years of
use. For women who had used COC for 15
years, the risk was almost halved. The pro-
tective effect started after at least 1 year
of COC use. The RR for < 1 year, 1–4 years,
5–9 years, 10–14 years, and ≥ 15 years of
use were 1.0, 0.78, 0.64, 0.56, and 0.42,
respectively.26
This reanalysis also found that the
more recent the use of COC, the lower the
RR of ovarian cancer. The proportional de-
cline in RR per 5 years of COC use was
29% for < 10 years of cessation of use,
19% for 10–19 years, and 15% for 20–29
years. The longer the duration of use, the
higher the protective effect irrespective
of the time elapsed from ceasing. The
start age of COC use and age of last use
seemed to have no effect on the protec-
tion. Low-dose pill use was found to have
an identical RR compared with high-dose
pill use.26
The Oxford-FPA contraceptive study,
which included 17,032 women aged 25–39
recruited at 17 family planning clinics in
England and Scotland between 1968–1974
with a long follow-up till 2004, analyzed
a total of 540,000 woman-years of ob-
servation and 58 ovarian cancer cases in
the control group and 48 cases in the COC
group.25 The overall ovarian cancer RR for
ever-users versus never-users was 0.5
(95% CI, 0.3–0.7). The risk of ovarian can-
cer was significantly lower in women on
oral contraceptives for more than 4 years.
The Royal College of General Prac-
titioners oral contraceptive study, which
started in 1968, collected data from
23,377 COC users and 23,796 never-users
over a period of 14 months, with 339,000
woman-years of observation for never-
users and 744,000 woman-years for ever-
users.17 It was found that the RR was 0.51
for ever-users as compared with never-
users. Similarly, a statistically significant
gradual decrease in risk with increasing
duration of COC use was observed. The
protective effect was found to last for at
least 15 years after stopping COC.
In the early 1970s, it was proposed
that defective cellular repair after ovula-
tion represents the major risk factor for
ovarian cancer development.27 More re-
cent theories assume that ovarian can-
cer development is attributed to either
activated proto-oncogenes or inactivated
tumour-suppressor genes,27,28 which seem
to point to abnormalities of genomic DNA
quantity and quality, with the resulting de-
fects in post-ovulatory ovarian cellular re-
pair being the causative factor for ovarian
cancer. Thus, the protective effect of COC
against ovarian cancer may be attributed
to the resulting anovulation during their
use, which prevents genetic predisposing
cellular repair defects to be expressed.
Although depot MPA also suppresses
ovulation and would theoretically lower
the risk of ovarian cancer, a hospital-based
WHO case-control study failed to uncover
such a protective effect.22
CARCINOMA OF THE CERVIX
There is evidence suggesting that long-
term use of COC for 5 years or more may
be associated with an increased risk of
cervical cancer.29 A meta-analysis of 28
studies, involving 12,531 women with
cervical cancer, suggested that the risk of
cervical cancer may decrease after stop-
ping the use of COC.30 Another IARC analy-
sis which included eight studies found a
fourfold increase in risk among women
with over 5 years of COC use. The risk was
also increased in women who started us-
ing COC before the age of 20 and in those
who had used COC within the previous 5
years.31
The mechanism for increased risk of
cervical cancer in COC users is uncertain.
Human papillomavirus (HPV) has been
recognized to be the major cause of carci-
noma of the cervix. Steroid contraception
has been postulated to be able to bind to
specific DNA sequences within transcrip-
tional regulatory regions on the HPV DNA,
either to increase or suppress the tran-
The protective effect of
COC against ovarian cancer
may be attributed to the
resulting anovulation
during their use
JPOG_JulAug_2013_CME_PH_Final_Hormonal Contraception and Cancers.indd 173 8/5/13 2:13 PM
JPOG JUL/AUG 2013 • 174
scription of various genes. It was suggest-
ed that the regulatory region of HPV type
16 viral genome indicates transcriptional
control of the HPV genome and might con-
tain enhancer elements that are activated
by steroid hormones.32
In COC users, the cervical mucus be-
comes scanty, thick, and highly viscous. It
has been hypothesized that such mucus
may modulate and prolong the effect of
carcinogenic agents and pathogens (in-
cluding HPV), which might have been car-
ried by coitus, on the cervical squamoco-
lumnar junction, causing them to become
difficult to be removed.33
However, the majority of studies did
not analyse the HPV status of COC users
and controls. Moreover, early use of COC
might be related to early onset of sexual
activity, which is itself a significant risk
factor for HPV infection and development
of cervical cancer. The lower use of the
barrier method of contraception in COC
users might be another accountable fac-
tor for the increase in risk of HPV infec-
tion. However, with the development of
HPV vaccines, the observed association
of increased risk of cervical cancer in COC
users might be changed, and fear of cervi-
cal cancer should not be a reason to avoid
COC use.
A large, population-based, case-con-
trol study in Costa Rica, a hospital-based
WHO case-control study in Thailand, Mex-
ico and Kenya, and a study in New Zealand
found that the risk of cervical cancer did
not appear to be affected by depot MPA
use.22
CONCLUSION
The majority of studies on the relationship
between hormonal contraceptive use and
development of cancer have focused on
COC and breast cancer, albeit with con-
flicting results. From the cumulative expe-
rience and meta-analyses of large epide-
miological studies with a long follow-up
duration, the present evidence suggests
an increase in risk of breast cancer devel-
opment mainly in current COC users, with
Figure 1. Risks of cancer development with duration of oral contraceptive pill (OCP) use
10
1.0
0.1
Rela
tive
risk
for c
ance
r dev
elop
men
t
Duration of OCP use (mo)0–48 49–96 > 97
Increased risk (?) of cervical cancer(Decreased risk on stopping OCP)
Increased risk of breast cancer(Decreased risk on stopping OCP)
Decreased risk of ovary cancer
Decreased risk of endometrial cancer
JPOG_JulAug_2013_CME_PH_Final_Hormonal Contraception and Cancers.indd 174 8/5/13 2:13 PM
Continuing Medical Education
JPOG JUL/AUG 2013 • 175
REFERENCES
the risk decreased on stopping therapy.
By 10 years of cessation of use, the risk
is similar to that in never-users. Similar-
ly, the risk of carcinoma of the cervix is
found to be increased in current COC us-
ers. However, COC use confers a strong
and prolonged protective effect against
carcinoma of the endometrium and ovary,
which will last even after over 20 years of
cessation of therapy (Figure 1).
Depot MPA was observed to cause
a plausible increased risk of breast can-
cer in current or recent users, similar to
that observed in COC users. It does not
seem to affect the overall breast cancer
risk. However, there is strong evidence of
prolonged decreased risk for carcinoma of
the uterine corpus. No strong association
has been noted for carcinoma of the cervix
and ovary.
When counselling women regarding
hormonal contraception, the issue of po-
tential carcinogenic effect from its use is
to be included. However, its prescription
should be based on an individual risk-
benefit assessment, provided contraindi-
cations are taken into account and regular
visits to doctors or health-care profession-
als are made.
1. Kirschstein RL, Rabson AS, Rusten GW. Infil-trating duct carcinoma of the mammary gland of a rhesus monkey after administration of an oral contraceptive: a preliminary report. J Natl Cancer Inst 1972;48:551–553.
2. Welsch CW, Adams C, Lambrecht LK, Hassett CC, Brooks CL. 17β-oestradiol and Enovid mam-mary tumorigenesis in C3H/HeJ female mice: counteraction by concurrent 2-bromo-alpha-ergocryptine. Br J Cancer 1977;35:322–328.
3. Geil RG, Lamar JK. FDA studies of estrogen, progestogens and estrogen/progesterone combi-nations in the dog and monkey. J Toxicol Environ Health 1977;3:179–193.
4. Shubik P. Oral contraceptives and breast cancer: laboratory evidence. IARC Sci Publ 1985;(65):33–35.
5. Anderson TJ, Battersby S, King RJ, McPher-son K, Going JJ. Oral contraceptive use influ-ences resting breast proliferation. Hum Pathol 1989;20:1139–1144.
6. Cogliano V, Grosse Y, Baan R, et al; WHO Inter-national Agency for Research on Cancer. Carci-nogenicity of combined oestrogen-progestagen contraceptives and menopausal treatment. Lan-cet Oncol 2005;6:552–553.
7. Preston-Martin S, Pike MC, Ross RK, Hender-son BE. Epidemiologic evidence for the increased cell proliferation model of carcinogenesis. Prog Clin Biol Res 1991;369:21–34.
8. Henderson BE, Feigelson HS. Hormonal car-cinogenesis. Carcinogenesis 2000;21:427–433.
9. Pike MC, Henderson BE, Casagrande JT, Rosa-rio I, Gray GE. Oral contraceptive use and early
abortion as risk factors for breast cancer in young women. Br J Cancer 1981;43:72–76.
10. Chilvers C, McPherson K, Peto J, et al; UK Na-tional Case-Control Study Group. Oral contracep-tive use and breast cancer risk in young women. Lancet 1989;333:974–982.
11. Cancer and Steroid Hormone Study of the Centers for Disease Control and the National In-stitute of Child Health and Human Development. Oral contraceptive use and the risk of breast can-cer. N Engl J Med 1986;315:405–411.
12. Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormonal con-traceptives: collaborative reanalysis of individual data on 53297 women with breast cancer and 100239 women without breast cancer from 54 epidemiological studies. Lancet 1996;347:1713–1727.
13. Colditz GA, Manson JE, Hankinson SE. The Nurses’ Health Study: 20-year contribution to the understanding of health among women. J Wom-ens Health 1997;6:49–62.
14. Kumle M, Weiderpass E, Braaten T, Persson I, Adami HO, Lund E. Use of oral contraceptives and breast cancer risk: The Norwegian-Swedish Women’s Lifestyle and Health Cohort Study. Cancer Epidemiol Biomarkers Prev 2002;11:1375–1381.
15. Vessey MP, McPherson K, Doll R. Breast can-cer and oral contraceptives: findings in Oxford-Family Planning Association contraceptive study. Br Med J (Clin Res Ed) 1981;282:2093–2094.
16. Kahlenborn C, Modugno F, Potter DM, Severs WB. Oral contraceptive use as a risk factor for premenopausal breast cancer: a meta-analysis.
Mayo Clin Proc 2006;81:1290–1302.
17. Hannaford PC, Selvaraj S, Elliott AM. Cancer risk among users of oral contraceptives: cohort data from the Royal College of General Practition-er’s oral contraception study. BMJ 2007;335:651.
18. Schneider HPG, Mueck AO, Kuhl H. IARC monographs program on carcinogenicity of com-bined hormonal contraceptives and menopausal therapy. Climacteric 2005;8:311–316.
19. Narod SA, Dubé MP, Klijn J, et al. Oral contra-ceptives and the risk of breast cancer in BRCA1 and BRCA2 mutation carriers. J Natl Cancer Inst 2002;94:1773–1779.
20. Paul C, Skegg DCG, Spears GFS. Depot me-droxyprogesterone (Depo-Provera) and risk of breast cancer. BMJ 1989;299:759–762.
21. Breast cancer and depot-medroxyprogester-one acetate: a multinational study. WHO Collabo-rative Study of Neoplasia and Steroid Contracep-tives. Lancet 1991;338:833–838.
22. Kaunitz AM. Depot medroxyprogesterone acetate contraception and the risk of breast and gynecologic cancer. J Reprod Med 1996;41(5 Suppl):419–427.
23. Backman T, Rauramo I, Jaakkola K, et al. Use of the levonorgestrel-releasing intrauter-ine system and breast cancer. Obstet Gynecol 2005;106:813–817.
24. Schlesselman JJ. Risk of endometrial cancer in relation to use of combined oral contraceptives: a practitioner’s guide to meta-analysis. Hum Re-prod 1997;12:1851–1863.
25. Vessey M, Painter R. Oral contraceptive use and cancer: findings in a large cohort study,
1968–2004. Br J Cancer 2006;95:385–389.
26. Collaborative Group on Epidemiological Stud-ies of Ovarian Cancer. Ovarian cancer and oral contraceptives: collaborative reanalysis of data from 45 epidemiological studies including 23,257 women with ovarian cancer and 87,303 controls. Lancet 2008;371:303–314.
27. Fathalla MF. Incessant ovulation—a factor in ovarian neoplasia? Lancet 1971;298:163.
28. Casagrande JT, Louie EW, Pike MC, Roy S, Ross RK, Henderson BE. “Incessant ovulation” and ovarian cancer. Lancet 1979;314:170–173.
29. Franceschi S. The IARC commitment to can-cer prevention: the example of papillomavirus and cervical cancer. Recent Results Cancer Res 2005;166:277–297.
30. Smith JS, Green J, Berrington de Gonzalez A, et al. Cervical cancer and use of hormonal contraceptives: a systematic review. Lancet 2003;361:1159–1167.
31. Moreno V, Bosch FX, Munoz N, et al . Effect of oral contraceptives on risk of cervical cancer in women with human papillomavirus infection: the IARC multicentric case-control study. Lancet 2002;359:1085–1092.
32. Moodley M, Moodley J, Chetty R, Herrington CS. The role of steroid contraceptive hormones in the pathogenesis of invasive cervical cancer: a review. Int J Gynecol Cancer 2003;13:103–110.
33. Guven S, Kart C, Guvendag Guven ES, Gu-nalp GS. The underlying cause of cervical can-cer in oral contraceptive users may be related to cervical mucus changes. Med Hypotheses 2007;69:550–552.
About the Author
Dr Wong is Consultant in the Department of Obstetrics and Gynaecology, Kwong Wah Hospital, Hong Kong.
JPOG_JulAug_2013_CME_PH_Final_Hormonal Contraception and Cancers.indd 175 8/5/13 2:13 PM
This continuing medical education service is brought to you by MIMS. Read the article ‘Hormonal Contraception and Cancers’ and answer the following questions. Answers are shown at the bottom of this page. We hope you enjoy learning with JPOG.
CME Answers for JPOG Jan/Feb 2012
HKCOG CME Article: Management of Pregnancies With Previous Caesarean Section
Answers
1 2 3 4 5 6 7 8 9 10
T F F T T T T T T F
CME Article
Hormonal Contraception and Cancers
Answer True or False to the questions below.
1. Combined oral contraceptives (COC) are classified by IARC as ‘carcinogenic’ to humans.
2. There is strong evidence for COC to cause a rise in epithelial cell proliferation of the glandular breast, leading to increase in genetic error and resulting in malignant transformation and breast cancer development.
3. The results of COC toxicity studies using animal models can accurately reflect the effects in humans.
4. The Collaborative Group on Hormonal Factors in Breast Cancer concluded that there was a small increase in the relative risk of having breast cancer in current and recent COC users.
5. Depot medroxyprogesterone acetate (MPA) users have decreased risk of carcinoma of the breast.
6. The protective effect of COC against carcinoma of the uterine corpus disappears after 10 years of stopping.
7. Low-dose COC has been shown to have a similar protective effect against carcinoma of the ovary compared with high-dose COC.
8. There is good evidence for the protective effect of depot MPA against carcinoma of the ovary.
9. There is evidence suggesting that long-term use of COC increases the risk of cervical cancer.
10. The carcinogenic potential of hormonal therapy should not be discussed during contraceptive counselling.
True False
JPOG JUL/AUG 2013 • 176
CME Questions CME QuestionsAnswers
1 2 3 4 5 6 7 8 9 10
T F F T F F T F T F
JPOG_JulAug_2013_CME_PH_Final_Hormonal Contraception and Cancers.indd 176 8/5/13 2:13 PM
Journal Watch
JPOG JUL/AUG 2013 • 133
Peer reviewed
GYNAECOLOGY
Calcium intake and mortality in Swedish women
Meta-analyses of randomized studies have shown
that taking calcium supplements is associated with
increased risk of coronary disease and stroke. A
Swedish cohort study has confirmed the increased
risk for cardiovascular disease in general but not
for stroke.
The Swedish mammography cohort was set
up in 1987 and included 61,433 women born be-
tween 1914 and 1948. National registries provided
data about all-cause and cardiovascular mortality
over a mean follow-up of 19 years. Food frequency
questionnaires in 1987 and 1997 provided data
about dietary intake and use of calcium supple-
ments for 38,984 women. The relationship between
calcium intake and all-cause mortality took the
form of a ‘J-shaped curve’, with higher mortality
at both extremes of intake. An intake of 1,400 mg
a day of calcium was associated with significant
increases of 40% in all-cause mortality, 49% in
cardiovascular mortality, and 114% in coronary
disease mortality, and no significant change in
stroke mortality, compared with a calcium intake of
600–1,000 mg a day. After further statistical analy-
sis, low intakes of calcium (< 600 mg/day) were
no longer significantly associated with increased
mortality. Among people taking calcium tablets and
with a dietary calcium intake of > 1,400 mg/day,
all-cause mortality was increased 2.6-fold.
High-calcium intake is associated with in-
creased all-cause and cardiovascular mortality.
Michaëlson K et al. Long-term calcium intake and rate of all cause and cardiovascular mortality: community-based prospective longitudinal cohort study. BMJ 2013; 346: 14 (f228).
Outcome of extreme preterm birth in England, 1995–2006
Two successive papers in the BMJ have examined
short-term and long-term outcomes for extremely
preterm births in England in 1995 and 2006.
The prospective national cohort studies pro-
vided short-term data about 666 babies born at 22–
25 weeks’ gestation in England in March to Decem-
ber 1995 and all 3,133 babies born at 22–26 weeks’
gestation in 2006. In 2006, 56% of infants born at
22 weeks and 98% of those born at 26 weeks were
born alive. Active care at birth was withheld from
73% of infants born at 22 weeks, 16% at 23 weeks,
and < 2% at 24 weeks or later. Survival rates for
live-born infants were 2% at 22 weeks, 19% at 23
weeks, 40% at 24 weeks, 66% at 25 weeks, and
77% at 26 weeks. More than two-thirds (68%) of
survivors had bronchopulmonary dysplasia, 16%
were treated (laser treatment) for retinopathy of
prematurity, and 13% had a serious abnormality on
cerebral ultrasound. In 2006, compared with 1995,
44% more infants born alive at 22–25 weeks were
admitted to neonatal intensive care, and survival
of infants born at 23, 24, and 25 weeks increased
by 9.5%, 12%, and 16%, respectively. Overall, the
proportion treated for retinopathy increased from
12% in 1995 to 22% in 2006.
Neurodevelopmental outcomes at ages 2–3
years were assessed for 1,031 survivors in the 2006
cohort. The prevalence of moderate or severe im-
pairment was 45% among survivors born at 22–23
weeks, 30% at 24 weeks, 25% at 25 weeks, and
20% at 26 weeks. Overall, one in seven (14%) had
cerebral palsy, usually mild or moderate. Mean
predicted adjusted mental development index quo-
tients (Bayley scales) were 80 (22–23 weeks), 87
(24 weeks), 88 (25 weeks), and 91 (26 weeks). In
the 2006 cohort, a greater absolute number of chil-
dren than in 1995 will need lifelong special care. It
is calculated that of every 100 infants born at 24
weeks, 60 will die despite intensive care and 12
of the 40 survivors will have serious impairments.
Further follow-up of the 2006 cohort is planned.
Between 1995 and 2006 the survival of ex-
tremely preterm infants in England improved, but
PAEDIATRICS
JPOG_JulAug_2013_COMBINE_Final.indd 133 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 134
more survived with disability. The writers of an
editorial point out that in the Netherlands, infants
born before 24 completed weeks are not routinely
offered neonatal intensive care.
Costeloe KL et al. Short term outcomes after extreme preterm birth in England: comparison of two birth cohorts in 1995 and 2006 (the EPICure studies). BMJ 2012; 345: 14 (e7976); Moore T. Neurological and developmental outcome in extremely preterm children born in England in 2006 and 1995: the EPICure studies. Ibid: 15 (e7961), Groenendaal F, Uiterwaal C. Long-term follow-up of extremely preterm neonates. Ibid:10 (e8252) (editorial).
New anti-interleukin therapies for systemic JIA
Systemic juvenile idiopathic arthritis (systemic JIA)
frequently leads to joint damage and disability and
is accompanied by systemic features such as fever,
rash, hepatosplenomegaly, and serositis. High-
dose steroid treatment may lead to severe toxicity,
and treatments such as methotrexate and tumour
necrosis factor inhibitors may be ineffective. The
effectiveness of antibodies to the interleukin-6
receptor and to interleukin-1β has been shown in
successive papers in the New England Journal of
Medicine.
The humanized, antihuman interleukin-6-re-
ceptor monoclonal antibody, tocilizumab, was as-
sessed in a placebo-controlled trial at 43 centres
in Europe North America, and South America. The
trial included 112 children aged 2–17 years with
active, treatment-resistant systemic JIA. Random-
ization (2:1) was to intravenous tocilizumab or
placebo every 2 weeks for 12 weeks. At 12 weeks,
an improvement of at least 30% on the American
College of Rheumatology JIA score (JIA ACR 30)
was achieved by 85% (tocilizumab) vs 24% (pla-
cebo), a highly significant difference. At week 52
a JIA ACR 70 response (at least 70% improvement)
was achieved by 80% in the tocilizumab group and
a JIA ACR 90 response by 59%. Steroid therapy had
been stopped by 52% in this group, and 48% had no
active arthritis. Common adverse events with tocili-
zumab included infections, neutropenia, and raised
aminotransferase levels.
The fully human, anti-interleukin-1β mono-
clonal antibody, canakinumab, was assessed in two
international trials reported together, including 84
and 100 patients. In the first trial, randomization
was to subcutaneous canakinumab or placebo, and
an adapted JIA ACR 30 response was achieved by
84% (canakinumab) vs 10% (placebo). In the sec-
ond trial, 100 patients who had responded to 32
weeks of canakinumab were randomized to con-
tinued canakinumab or to placebo. A disease flare
occurred in 74% (canakinumab) vs 25% (placebo).
The median time to disease flare was incalculable
in the canakinumab group and 236 days in the pla-
cebo group. The disease became inactive in 62%
vs 34%. One in three patients on canakinumab was
able to discontinue steroid therapy. Infections were
frequent with canakinumab and five patients (ver-
sus two in the placebo group) developed the mac-
rophage activation syndrome.
Both tocilizumab and canakinumab were ef-
fective treatment for systemic JIA but more data
are needed about toxicity.
De Benedetti F et al. Randomized trial of tocilizumab in systemic juvenile idiopathic arthritis. NEJM 2012; 367: 2385–95; Ruperto N et al. Two randomized trials of canakinumab in systemic juvenile idiopathic arthritis. Ibid: 2396–406; Sandborg C, Mellins ED. A new era in the treatment of systemic juvenile idiopathic arthritis. Ibid: 2439–40 (editorial).
CPAP vs surfactant and higher vs lower oxygen saturation for extremely preterm infants: Outcomes at 18–22 months
The Surfactant, Positive Pressure, and Pulse Oxim-
etry randomized trial was a multicentre, random-
ized, controlled trial with a 2 × 2 multifactorial
design in which 1,316 extremely preterm infants
(born at 24 weeks 0 days to 27 weeks 6 days) were
randomized at 20 US centres to early continuous
positive airway pressure (CPAP), or early surfactant
via an endotracheal tube and to a target oxygen
saturation of 85–89% or of 91–95%. Early assess-
ment (at 36 weeks’ postmenstrual age) showed
similar rates of death or bronchopulmonary dyspla-
sia with either CPAP or surfactant, and the lower
target range for oxygen saturation was associ-
ated with less retinopathy of prematurity but more
JPOG_JulAug_2013_COMBINE_Final.indd 134 8/5/13 2:18 PM
Journal Watch
JPOG JUL/AUG 2013 • 135
Peer reviewedPeer reviewed
deaths. Now, surviving infants have been assessed
at 18–22 months.
Neurodevelopmental status was assessed
at 18–22 months in 990 of 1,058 surviving infants
(94%). Death or neurodevelopmental impairment
occurred in 27.9% (CPAP) vs 29.9% (surfactant),
a non-significant difference, and in 30.2% (lower
oxygen saturation target) vs 27.5% (higher oxygen
saturation target), also a non-significant differ-
ence. There was a significant increase in mortality
with the lower oxygen saturation target (22.1% vs
18.2%).
These researchers conclude that outcomes
are similar with early CPAP and with early surfac-
tant, but the lower oxygen saturation target should
not be used in the care of extremely premature
babies.
Vaucher YE et al. Neurodevelopmental outcomes in the early CPAP and pulse oximetry trial. NEJM 2012; 367: 2495–504.
Paromomycin for cutaneous leishmaniasis
York. It included inpatients and outpatients in No-
vember to May each year between 2003 and 2009
with children presenting with an acute respiratory
illness or fever. The rate of HMPV detection (using
reverse transcriptase–polymerase chain reaction
assay on nose and throat swabs) was 200/3,490
(6%) among children in hospital, 222/3,257 (7%)
among children in outpatient clinics, 224/3,001
(7%) among children in the emergency department,
and 10/770 (1%) among healthy children in well-
child primary care clinics. Rates of hospital admis-
sion with HMPV infection among children aged < 5
years were 1 in 1,000 (3 in 1,000 at age < 6 months,
and 2 in 1,000 at age 6–11 months). Among chil-
dren admitted to hospital with an acute respiratory
illness or fever, those with HMPV infection were
older, more likely to be diagnosed as pneumonia or
asthma, to need supplemental oxygen, and to stay
longer in intensive care, compared with children
testing negative for HMPV. It was estimated that
among 1,000 children of this age, HMPV would,
each year, cause 55 clinic visits and 13 visits to the
emergency department. Among children admitted
to hospital, coexisting high-risk conditions (pre-
Leishmaniasis is prevalent in Eurasia, Africa, and
the Americas, and although cutaneous leishmani-
asis eventually resolves without treatment it is the
cause of much morbidity. Cutaneous leishmaniasis
due to Leishmania major is prevalent in Tunis, and
a trial there has shown topical paromomycin to be
effective treatment.
A total of 375 patients aged 5–65 years (half
of them children < 17 years old) were randomized
to three groups: 15% paromomycin, 15% paromo-
mycin plus 0.5% gentamicin, or vehicle alone (pla-
cebo), all applied as topical creams for 20 days to
all ulcerated skin lesions including an index lesion
(1–5 cm diameter with leishmania demonstrated).
Cure of the index lesion was achieved in 82% (par-
omomycin), 81% (paromomycin/gentamicin), and
58% (placebo). Only seven patients (five in the pla-
cebo group) had any persisting lesions after cure of
the index lesion. Mild to moderate local reactions
occurred with paromomycin.
Topical treatment with paromomycin cream,
with or without gentamicin, was effective treat-
ment for cutaneous leishmaniasis due to L major.
Salah AB et al. Topical paromomycin with or without gentamicin for cutaneous leishmaniasis. NEJM 2012; 368: 524–32.
HMPV infection in young children in the US
Human metapneumovirus (HMPV) is a paramyxovi-
rus discovered in 2001 as a cause of acute respi-
ratory illness in infants and young children world-
wide. It also affects old people and people with
debilitating illnesses. A study at three US sites
has provided more data about the epidemiology of
HMPV in children under the age of 5 years.
The survey by the Centers for Disease Con-
trol and Prevention (CDC) New Vaccine Surveillance
network took place at three sites, in Cincinnati,
Ohio; Nashville, Tennessee; and Rochester, New
JPOG_JulAug_2013_COMBINE_Final.indd 135 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 136
mature birth, immunodeficiency, chronic disease of
lungs, heart, or kidneys, cancer, or sickle-cell dis-
ease) were present in 40% (HMPV-positive) vs 30%
(HMPV-negative).
In the US, HMPV is an important cause of
acute respiratory illness and fever among young
children.
Edwards MK et al. Burden of human metapneumovirus infection in young children. NEJM 2013; 368: 633–43.
Risk factors for stillbirth
There has been little improvement in stillbirth rates
in recent years, and the importance of risk factors
has been uncertain. A study in the West Midlands
region of England has included 91,829 live births
and 389 stillbirths.
Risk factors for stillbirth included parity
(para 0 and 3+), ethnicity (African, Afro-Caribbean,
using country-specific data) for infants born in
hospital in 23 developing countries in Africa and
Latin America (2004–2005) and Asia (2007–2008),
including 276,436 singleton live births or fresh still-
births. The 90th percentile for birth weight varied
from 3,250 g in India to 4,050 g in Algeria, and the
prevalence of a birth weight of 4,000 g or greater
was 0.5% in India and 14.9% in Algeria. Factors
found to be significantly associated with macroso-
mia included higher maternal age (20–34 years),
higher maternal height, higher parity, higher mater-
nal BMI, maternal diabetes, post-term pregnancy,
and male fetus. Macrosomia was associated with
increased risk of caesarean section and of adverse
maternal outcomes. The risk of adverse perinatal
outcome was increased in Asia.
The increase in obesity and diabetes in
women of reproductive age might have led to an
increase in fetal macrosomia worldwide. Research
into ways of controlling these factors is needed.
Koyanagi A et al. Macrosomia in 23 developing countries: an analysis of a multicountry, facility-based, cross-sectional survey. Lancet 2013; 381: 476–83; Dennedy M, Dunne F. Macrosomia: defining the problem worldwide. Ibid: 435–6 (comment).
Indian, or Pakistani), maternal obesity (BMI 30 or
greater), smoking, pre-existing diabetes, history of
mental health problems, and pregnancy complica-
tions (antepartum haemorrhage, fetal growth re-
striction). The greatest risk factor was fetal growth
restriction which increased the risk of stillbirth by a
factor of 7.8 in non-smoking mothers, 5.7 in smok-
ing mothers, and 10.0 in mothers only exposed to
passive smoking. The population attributable risk
from fetal growth restriction was 6.2% if detected
antenatally and 32.0% when not detected antena-
tally. Antenatal detection of fetal growth restric-
tion was associated with delivery 10 days earlier
on average. The stillbirth rate (per 1,000 births)
was 4.2 overall, 2.4 in pregnancies with no fetal
growth restriction, 9.7 when fetal growth restric-
tion was detected antenatally, and 19.8 when fetal
growth restriction was not detected antenatally.
This study identifies fetal growth restriction
as the main risk factor for stillbirth. Antenatal de-
tection of fetal growth restriction and appropriate
early delivery might prevent 600 stillbirths each
year in the UK.
Gardosi J et al. Maternal and fetal risk factors for stillbirth: population based study. BMJ 2013; 346: 15 (f108); McCowan LME, Groom KM. Identifying risk factors for stillbirth. Ibid: 7 (f416) (editorial).
Fetal macrosomia in developing countries
Fetal macrosomia may lead to perinatal death,
perinatal asphyxia, shoulder dystocia, caesarean
section, maternal haemorrhage, prolonged labour,
and perinatal trauma. In the developed world, the
prevalence of macrosomia has increased along
with maternal obesity and diabetes. Little is
known, however, about fetal macrosomia in devel-
oping countries. Now, the WHO Global Survey on
Maternal and Perinatal Health has provided data
about macrosomia (birth weight > 90th percentile
OBSTETRICS
JPOG_JulAug_2013_COMBINE_Final.indd 136 8/5/13 2:18 PM
Imaging Paediatric Brain Tumours
Tang Phua Hwee, MBBS, FRCR, MMed Diagnostic Radiology
PAEDIATRICS i Peer reviewed
JPOG JUL/AUG 2013 • 137
INTRODUCTION
Prompt identification and management of hearing loss in childhood is essential to en-
sure optimal speech and language development in the early years of life and optimal
school performance for older children. The implementation of universal neonatal hear-
ing screening has facilitated earlier identification of congenital losses, with 1 in 1,000
babies being born in the UK with a permanent hearing loss detectable at birth. However,
for more common acquired otological conditions of childhood such as otitis media with
effusion (OME) or hearing loss in the presence of complex medical needs, the paediatri-
cian may be the first port of call for parents or concerned allied health professionals,
and an understanding of the presentation, assessment and management options can
ensure optimal outcomes for this group of children. The aim of this review is to discuss
the clinical assessment of hearing loss and provide an overview of the management
options available.
Types of Hearing LossHearing loss is classified by both type and severity. The types of hearing loss include
conductive, sensorineural, or mixed hearing loss, which can be subdivided into congeni-
tal or acquired. A differential diagnosis is included in Table 1. The severity of hearing
loss in decibels hearing level (dBHL) is based on a ‘pure tone average’, which is the
mean hearing threshold at four different sound frequencies (0.5, 1, 2, and 4 kHz). This
allows classification of the patients hearing as normal (< 25 dBHL) or determines the
extent of the loss; mild (25–50 dBHL), moderate (51–70 dBHL), severe (71–90 dBHL),
profound (91–110 dBHL), or total (> 110 dBHL). Management of the hearing loss thus
depends on its aetiology and the extent of the loss.
PAEDIATRICS i Peer reviewed
The Management of
Hearing Loss in Children
Marianne D Elloy, MBBS, MRCS, DOH-NS; Andrew H Marshall, FRCS(ORL-HNS)
JPOG_JulAug_2013_COMBINE_Final.indd 137 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 138
PAEDIATRICSPAEDIATRICS I Peer revIewed
Table 1. Differential diagnosis of paediatric hearing loss
Inheritance/aetiology Sensorineural HL Conductive HL
Congenital Non-syndromic
Autosomal dominantAutosomal recessiveX-linked
DFNA chromosome loci: approx 40 gene loci identified. Most common is GJB2 (connexin 26)
DFNB chromosome loci: approx 35 gene loci identified. Most common is GJB2 (connexin 26)
DFN chromosome loci: approx 5 gene loci identified
Congenital ossicular anomalies/fixation
Syndromic Autosomal Dominant Branchio-oto-renal syndromeCraniosynostosis syndromes (Crouzon, Apert,
Muenke, Pfeiffer)DiGeorge syndromeNoonan syndrome
Osteogenesis imperfectaMarshall syndromeNeurofibromatosis type 2Saethre-Chotzen syndrome Stickler syndromeWaardenburg syndrome
Treacher Collins syndrome
Autosomal recessive Albers-Schonberg diseasePendred syndromeEnlarged vestibular aqueduct
syndromeUsher syndrome
X-linked Alport syndromeMohr-Tranebjaerg syndrome
Trisomy Sporadic, autosomal dominant or autosomal recessive
Down syndromeCHARGE syndrome
Goldenhar syndrome
Alport syndrome
Acquired Prenatal Intrauterine infectionsIntrauterine exposure to
ototoxic drugsPostnatal Infectious AOM and complications of
AOM MeningitisViral: measles, mumps, CMV
Inflammatory Chronic suppurative otitis media
OME
Trauma Tympanic membrane perforation
Temporal bone fractureAOM = acute otitis media; CMV = cytomegalovirus; OME = otitis media with effusion.
JPOG_JulAug_2013_COMBINE_Final.indd 138 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 139
PAEDIATRICSPAEDIATRICS I Peer revIewed
CLINICAL ASSESSMENT
detectionThe detection of hearing loss either occurs as a re-
sult of a universal neonatal hearing screening pro-
gramme or as a result of parental or professional
concern.
Universal neonatal hearing screening is now
well established in the United Kingdom, and its aim
is for early identification of hearing loss as early
intervention results in better speech and language
outcomes.
The programme aims for babies to be screened
by the gestational age of 44 weeks but preferably
before discharge from hospital. Premature babies
should not be screened before the gestational age
of 34 weeks. Two screening pathways exist: the
well baby protocol and the neonatal intensive care
unit/special care baby unit (NICU/SCBU) protocol
(for those babies admitted to NICU/SCBU for more
than 48 hours). The exclusions for hearing screen-
ing are ear atresia, microtia, or meningitis, with a
recommendation of direct referral for audiological
assessment in this high-risk group.
A dedicated hearing screening team under-
takes the assessments. The first step of the well
baby protocol comprises otoacoustic emissions
(OAEs) testing. If the baby passes this test and
there is no history of risk factors, then the baby is
discharged from the programme. If the baby fails
the test, it can be repeated on a second occasion;
however, if the test is failed again, an automated
auditory brainstem response (AABR) test is per-
formed. If this is passed and there are no risk fac-
tors, the baby can be discharged; however, if the
AABR is missed or incomplete, or if there are no
clear responses in one or both ears, or risk factors
are identified, further audiological assessment will
be arranged. The NICU/SCBU protocol differs with
all babies undergoing an OAE and AABR test. The
audiological referral criteria are the same as for
the well baby protocol; in addition, if either test
is missed or the tests provide inconsistent results,
then audiological review is recommended. A poten-
tial pitfall using OAEs as a screening tool is the po-
tential failure to identify a condition called auditory
neuropathy. In these children, OAEs are present but
their auditory brainstem response and functional
hearing can be very poor. This condition is more
prevalent in babies that would be identified by the
NICU/SCBU protocol in particular due to prematu-
rity or neonatal infection.
HistoryThe history in neonates should explore the risk fac-
tors identified in the neonatal hearing screening
pathway which includes parental or professional
concern regarding the infants’ hearing or develop-
ment of auditory or vocal behaviour, high risk of
chronic middle ear problems, for example, Down
syndrome or cleft palate, craniofacial anomalies,
family history of permanent sensorineural hearing
loss from early childhood (in parents or siblings),
intermittent positive pressure ventilation, on NICU
or SCBU for more than 5 days, jaundice or hyperbili-
rubinaemia requiring exchange transfusion, proven
or possible congenital infections, TORCH (Toxoplas-
mosis Other: syphilis Rubella CMV Herpes), neuro-
degenerative or neurodevelopmental disorders, and
exposure to ototoxic drugs with monitored levels
outside the therapeutic range. However, it is im-
portant to be aware that 50% of newborns born
with permanent bilateral congenital hearing loss
do not have any known risk factors. Other risk fac-
tors which have been identified include bacterial
meningitis, whereby 10% of children will develop
a subsequent sensorineural hearing loss, Apgar
scores (0–4 at 1 minute and 0–6 at 5 minute), birth
weight less than 1,500 g, and consanguinity.
The history in older children should focus on
JPOG_JulAug_2013_COMBINE_Final.indd 139 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 140
PAEDIATRICSPAEDIATRICS I Peer revIewed
the impact of the hearing loss on the child’s speech
and language development, and school perfor-
mance including social interaction with their peers.
The aforementioned risk factors should be explored
and, in addition, enquiry about previous head injury
and temporal bone fracture.
examinationIn neonates with hearing loss, a full clinical exami-
nation should be undertaken as syndromic hearing
loss can occur in up to 30% of children with bilat-
eral permanent hearing loss and other associated
features may be noted.
Otoscopy may be normal. Examination for
craniofacial anomalies, preauricular pits, sinuses,
and branchial pits should be undertaken. Some syn-
dromes with hearing loss can be associated with
ophthalmological anomalies and all children with
a moderate or greater sensorineural hearing loss
should be reviewed by an ophthalmologist.
In older children, the most common cause of
childhood hearing loss is OME, and a number of dif-
ferent characteristic otoscopic appearances have
been identified including a dull tympanic membrane,
loss of the light reflex, flattening of the handle of
the malleus, a golden or blue hue of the tympanic
membrane, and indeed sometimes an air fluid me-
niscus or retrotympanic bubbles. Otosocopy should
also assess for the presence impacted cerumen, a
foreign body, infection, congenital cholesteatoma,
tympanosclerosis (the presence of calcification in
the tympanic membrane which can sometimes also
affect the ossicular chain), perforation (including
the site, extent, and status of the middle ear mu-
cosa), or an attic defect with the possibility of cho-
lesteatoma.
The facial nerve should also be assessed par-
ticularly in the presence of pathology.
investigations
Audiological Assessments
The objective audiological assessments can be per-
formed for children of any age as no contribution to
the testing process is required by the patient.
Otoacoustic emissions: the principle of
OAEs is that objective sounds are emitted from the
outer hair cells of a normally functioning cochlea.
OAEs can be spontaneous or occur in response to
acoustic stimulus. Transient-evoked OAEs are used
in neonatal hearing screening, whereby broadband
clicks are delivered to the ear by a handheld probe
which also contains a microphone to detect the
emissions. The presence of OAEs indicates a hear-
ing threshold of 20–40 dBHL. The test is quick and
easy to administer, is not affected by sleep, and has
a high sensitivity (97%), making it a useful screen-
Syndromic hearing
loss can occur in up
to 30% of children
with bilateral
permanent hearing
loss and other
associated features
may be noted
JPOG_JulAug_2013_COMBINE_Final.indd 140 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 141
PAEDIATRICSPAEDIATRICS I Peer revIewed
ing tool. The limitations include inability to esti-
mate hearing thresholds or assess specific frequen-
cies. Specificity is low and patients can fail if they
have impacted wax or OME. In contrast, a patient
can pass this test but still have hearing problems
due to an auditory neuropathy.
Auditory brainstem response: Auditory
brainstem response can be undertaken as an auto-
mated test or a manually interpreted test to gain in-
formation about frequency-specific hearing thresh-
olds including bone conduction thresholds and
aided thresholds, which is useful in the presence
of ear canal atresia and microtia. The test takes
longer to administer and may require sedation or
general anaesthesia. The test records the activity
of the eighth cranial nerve and auditory pathways in
response to acoustic stimulus (via headphones) by
adhesive scalp electrodes. It has a high specificity
and sensitivity (> 90%).
Tympanometry: this test does not assess
hearing but is used to assess the compliance of
the tympanic membrane and is particularly useful
in the assessment of OME. The shape of the graph
produced gives information about middle ear com-
pliance, and a normal peak (type A) suggests nor-
mal middle ear function, a flattened peak (type B)
is suggestive of OME (or perforation in the presence
of a high ear canal volume), and a peak shifted to a
more negative pressure (type C) suggests Eustachi-
an tube dysfunction.
Behavioural Testing
Visual reinforcement audiometry: this can be
used for children age 6 months to 3 years. It is un-
dertaken in a specially adapted audiology room us-
ing either sound field speakers or ear inserts. The
child must be able to sit on a parent’s lap and be
able to turn their head to the sound. A distracter en-
tertains the child with toys and sounds are played
via the speakers; when the child turns correctly in
response to the sound, a visual reward is triggered
by the audiologist conducting the test, for example,
an illuminated moving toy. The advantages are that
hearing thresholds can be determined; however,
children can tire and lose interest which reduces
accuracy.
Conditioning/play audiometry: this can be
used for children aged above 2–3 years who can
obey simple commands. The child is instructed to
perform a task each time they hear a sound pre-
sented by headphones or sound field, for example,
put a wooden man into a wooden boat. This is easy
to perform and can establish hearing thresholds but
is dependent on the compliance of the child.
Pure tone audiometry: this is the same test
as used for testing the hearing in adults, requiring
the patient to press a button each time they hear
a sound. This can be used in children aged 4 years
and above, depending on their compliance. Parents
and siblings should be investigated with pure tone
audiometry.
Imaging: imaging is indicated in patients with
a bilateral severe to profound hearing loss, severe
to profound unilateral hearing loss, or a progres-
sive loss. Local protocols vary and a combination of
computed tomography or magnetic resonance imag-
ing (MRI) scanning can be utilized. The anatomical
features or variations of the cochlear nerves and
structure of the cochlea are well demonstrated by
MRI. However, the presence of ossification within
the cochlea, in particular, following meningitis is
optimally imaged using computed tomography.
Other investigationsElectrocardiogram (ECG): all children identified
to have a bilateral permanent hearing loss should
undergo an ECG to assess for a prolonged QT inter-
val which is typically found in Jervell and Lange-
Neilsen syndrome, and if anomalies are identified
referral to a cardiologist should be undertaken.
JPOG_JulAug_2013_COMBINE_Final.indd 141 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 142
PAEDIATRICSPAEDIATRICS I Peer revIewed
Blood tests and urinalysis: screening for
congenital infections can be undertaken by test-
ing for antibodies to toxoplasmosis, syphilis, ru-
bella, cytomegalovirus, and herpesvirus hominis. A
blood test for connexin 26, the most common non-
syndromic genetic cause of hearing loss, can be
undertaken. Urea and electrolytes accompanied by
urinalysis for haematuria should be undertaken as
part of a screen for Alport syndrome and branchio-
oto-renal syndrome. Thyroid function tests may be
normal in the early stages of Pendred syndrome,
and the classical investigation described is the per-
chlorate discharge test.
MANAGEMENT
For optimal outcomes, the management of paedi-
atric hearing loss requires a multidisciplinary ap-
proach. This can involve a host of medical speciali-
ties, in addition to ear, nose and throat surgeons,
the general practitioner, paediatricians, ophthal-
mologists, to manage ocular anomalies and the
geneticists to investigate aetiology of hearing loss
and undertake genetic counselling with the child’s
family.
The child’s family will require a significant
amount of support and input from the allied health
professionals. The audiologists not only assess
the child’s hearing thresholds but also are the key
professional group in the provision of adjuncts to
hearing rehabilitation including the management
of hearing aids, bone anchored hearing aids, and
cochlear implant programming. The teachers of
the deaf provide support in the learning environ-
ment before children even start school. They spend
time in the classroom to assess what adjuncts may
be required and train the teachers in mainstream
schools on how to support children with hearing im-
pairment. The speech and language therapists’ in-
put is essential to optimize language development,
and they often work both in the community and the
school to provide input. At the initial consultation,
simple measures to enhance a child’s hearing abili-
ties should be discussed with the parents, including
classroom placement strategies.
Conservative ManagementChildren who are making satisfactory progress de-
spite a mild to moderate hearing loss may be man-
aged with supportive measures. The majority of
children with OME are managed conservatively as a
significant proportion resolve spontaneously.
In those children requiring intervention, the
use of hearing aids for amplification is the main-
stay of auditory rehabilitation. Most children will
require behind-the-ear digital hearing aids, with
multiple brands and models available. These can be
programmed to the child’s hearing loss and a mould
can be made to fit specifically in their ear, which
can be personalized with the logo of their favourite
football team being popular for boys and glitter be-
ing a hit with girls! For children with ear canal atre-
Figure 1. Soft band hearing aid.
JPOG_JulAug_2013_COMBINE_Final.indd 142 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 143
PAEDIATRICSPAEDIATRICS I Peer revIewed
sia, microtia and other conductive deficits, a soft
band hearing aid – a bone conductor hearing aid
attached to a soft head band is particularly useful
(Figure 1). Adjunctive devices such as FM systems
for the classroom can also be valuable in hearing
rehabilitation.
Surgical ManagementGrommets (ventilation tubes): grommets are
small (plastic or titanium) tubes that are inserted
into the tympanic membrane to aerate the middle
ear. Multiple types of grommets are available with
the design impacting on the duration that the grom-
met is maintained in position. The main indication
for grommet insertion is persistent OME with hear-
ing loss. The National Institute for Clinical Excel-
lence (NICE) guidance (2008) advocates a 3-month
period of active observation as during this period
the effusion will resolve in 60% of children. For
persistent bilateral OME with hearing in the better
ear of 25–30 dBHL or worse, grommets can be of-
fered, or if the hearing is less than 25–30 dBHL but
is significantly impacting the child’s development
or education. Children with Down syndrome often
have persistent problems with OME, and grommet
placement can be technically difficult owing to nar-
row ear canals and children often require multiple
sets of grommets. As such, a separate pathway is
provided for children with Down syndrome, with
more emphasis on utilization of hearing aids than
grommets. A third pathway for children with cleft
palate advocates the use of grommets in the pres-
ence of persistent OME and hearing loss.
The advantages of grommet placement are
instantaneous improvement in hearing. This quick
procedure is carried out under general anaesthetic
in children and often preferred by parents to the
alternative; a hearing aid as compliance is not an
issue. The risks of surgery include bleeding, infec-
tion, and perforation. In the event of infection, a
1-week course of topical antibiotic ear drops is ad-
vocated as the first-line treatment, on rare occa-
sions, for refractory infections grommets have to be
surgically removed.
Bone-anchored hearing aid (BAHA): the
indications for BAHA in children include congeni-
tal aural atresia and microtia, chronic suppurative
otitis media, persistent OME, chronic otitis externa,
unilateral profound hearing loss, failure with con-
ventional aids, and trauma to the external ear ca-
nal. A BAHA is a titanium screw inserted into the
Figure 2. Cochlear implant.
How a cochlear implant works1. The sound processor (A) captures sound and converts it
into digital code.2. The sound processor transmits the digitally coded sound
through the coil (B) to the implant (C) just under the skin.3. The implant converts the digitally coded sound to electrical
signals and sends them along the electrical signals and sends them along the electrode array, which is positioned in the cochlea.
4. The implant’s electrodes stimulate the cochlea’s hearing nerve fibres, which relay the sound signals to the brain to produce hearing sensations.
JPOG_JulAug_2013_COMBINE_Final.indd 143 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 144
PAEDIATRICSPAEDIATRICS I Peer revIewed
calvarial bone behind the ear. Osseointegration
occurs, resulting in bony growth into the titanium
screw. Once this has occurred, typically after 3–6
months, a bone conduction hearing aid system can
be attached, providing the child with amplification.
Typically, children cannot undergo implantation un-
til the age of 4 years; until a child is old enough to
undergo implantation, a soft band hearing aid can
be used.
The risks of surgery include failure of osse-
ointegration, soft tissue complications, trauma,
fixture displacement, and extradural haematoma.
Outcomes measures include favourable audiologi-
cal outcomes in comparison to conventional bone
conductor aids and good patient usage and satis-
faction after up to 10 years after fitting.
Cochlear implant: A cochlear implant is
a surgically implantable device which bypasses
damaged hair cells in the cochlea and directly
stimulates the cochlear nerve (Figures 2 and 3). It
comprises a microphone (which captures speech
and sound), processor (which converts the sound
into an electronic signal), transmitting coil (which
transmits the electronic signal), internal receiver/
stimulator (which receives the electronic signal and
converts it to electrical impulses), and an electrode
array (which delivers the electrical impulses to the
cochlear nerve). Modern electrodes are multichan-
nel with the common brands having 16 and 22 chan-
nels. Each channel stimulates a specific frequency.
Insertion of an electrode into the cochlear risks
destroying any residual hearing and makes the ear
reliant on the cochlear implant. However, hybrid
cochlear implants have been recently developed
for use in patients with serviceable low-frequency
hearing but a profound high-frequency loss, where-
by a custom made shortened electrode is inserted
with the aim of preserving existing hearing in the
low frequencies.
NICE guidance introduced in 2009 recom-
mended that simultaneous bilateral cochlear im-
plantation was indicated in children with severe
to profound hearing loss who did not receive ad-
equate benefit from conventional hearing aids. The
guidelines defined the severe to profound hearing
loss as ‘only hearing sounds louder than 90 dBHL
at the frequencies of 2 and 4 kHz without hearing
aids’ and recommended the utilization of speech,
language, and listening skills appropriate for age,
Figure 3. Cochlear implant.
JPOG_JulAug_2013_COMBINE_Final.indd 144 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 145
PAEDIATRICSPAEDIATRICS I Peer revIewed
combined with development stage and cognitive
ability as factors to consider when assessing the
benefit of hearing aids. The main contraindications
to cochlear implantation include lack of a cochlear
nerve and complete ossification of the cochlea.
Patients who are candidates for cochlear im-
plantation will be assessed by a specialist multi-
disciplinary cochlear implant team including au-
diological investigations, imaging, trial of hearing
aids, review by the speech and language therapists,
and teachers of the deaf to assess for suitability.
Confirmation that children are vaccinated against
pneumococcus will be sought preoperatively from
the general practitioner.
The risks of surgery include altered taste, in-
fection, meningitis, device failure, and facial nerve
injury. It is also important to note that children with
cochlear implants cannot undergo MRI.
Outcome measures have been used to assess
the efficacy of cochlear implantation, and there is
a wealth of literature supporting the success of
hearing rehabilitation and these include ‘sound lo-
calization’ and ‘speech recognition in noise’ meas-
ures and quality of life. Reports have revealed that
many children who received cochlear implants at an
early age are achieving age-appropriate academic
performance.
FURTHER READING
Browning G. Clinical otology & audiology. London: Arnold, 2001.Gerber S. The handbook of pediatric audiology. Washington: Gallau-
det, 1996.Graham J, Scadding G, Bull P. Pediatric ENT. Heidelberg: Springer,
2007.Johnston J, Durieux-Smith A, O’Connor A, Fitzpatrick E. Bilateral
cochlear implants: a critical review. Int J Audiol 2009;48:601–617.Kral A, O’Donoghue G. Profound deafness in childhood. N Engl J Med
2010;363:1438–1450.Kunst D, Kremer H, Cremers C. Genetics for ENT specialists. London:
Remedica, 2005.McDermott A-L, Sheehan P. Bone anchored hearing aids in children.
Curr Opin Otolaryngol Head Neck Surg 2009;17:488–493.NHS. Newborn Hearing Screening Programme. Available online at
http://hearing.screening.nhs.uk/; 2011 (accessed 5 Mar 2011).NICE. Cochlear implants for children and adults with severe to
profound deafness: NICE technology appraisal guidance 166. Avail-able online at http://guidance.nice.org.uk/TA166; 2009 (accessed 5 Mar 2011).
NICE. Surgical management of otitis media with effusion in children: NICE clinical guideline 60. Available online at http://guidance.nice.org.uk/CG60; 2008 (accessed 5 Mar 2011).
Papsin B, Gordon K. Bilateral cochlear implants should be the stan-dard for children with bilateral sensorineural deafness. Curr Opin Otolaryngol Head Neck Surg 2008;16:69–74.
Phillips J, Yung M, Burton M, Swan I. Use of aminoglycoside contain-ing ear drops in the presence of a perforation evidence review and ENT UK consensus statement. Available online at http://www.entuk.org/news/news/attachments/eardrops; 2007 (accessed 5 Mar 2011).
US Preventative Services Task Force. Universal screening for hearing loss in newborns: US Preventative Services Task Force recommen-dation statement. Pediatrics 2008;122:143–148.
© 2011 Elsevier Ltd. Initially published in Paediatrics and Child Health 2011;22(1):13–18.
About the AuthorsMarianne D Elloy is Specialist Registrar in Otolaryngology at the Royal Derby Hospital, Derby, UK. Andrew H Marshall is Consul-tant in Otolaryngology at the Queens Medical Centre, Notting-ham, UK.
Practice points
• Universal neonatal hearing screening identifies 1 in 1,000 babies with a permanent hearing impairment at birth.
• Management of hearing loss is a multidisciplinary approach.• Hearing loss can be associated with a syndrome, and associated
features should be identified and investigated.• Early identification and intervention of hearing loss improves
speech and language outcomes.• Conventional digital hearing aids or soft band bone conductor
hearing aids can be used for children.• If hearing aids provide inadequate response to sound, patients
can be referred for assessment for either bone-anchored hearing aid or cochlear implants, depending on their hearing thresholds and pathology.
JPOG_JulAug_2013_COMBINE_Final.indd 145 8/5/13 2:18 PM
Professor Nimish Vakil talks about management of patients with refractory GERD.“Successful treatment of refractory GERD requires thorough investigation of the patient situation.”
Professor David Liebermanshares his perspective on the present and future of colorectal cancer screening.“There is a lot of potential to prevent many cancers if we can improve the rate of CRC screening.”
Dr Markus Cornbergdiscusses the management of chronic hepatitis B.“The aim of therapy should be the cure or control of HBV infection without the need for life-long treatment.”
In this Series, find out what these medical experts have to say about latest updates in the management of refractory GERD, the management of chronic hepatitis B and the present & future of colorectal cancer screening.
Current Opinion in Gastroenterology
SCAN TO WATCH VIDEO
Brought to you by MIMS
MIMS Video Series features interviews with leading experts.
Got a spare 5 minutes? Go to www.mims.asia/video_series
MIMS Video Series
By Doctors • For DoctorsBY DOCTORS
FOR DOCTORS
MIMSVideo2013_GERD_ad_206x276.indd 1 1/8/13 12:42 PM
Imaging Paediatric Brain Tumours
Tang Phua Hwee, MBBS, FRCR, MMed Diagnostic Radiology
PAEDIATRICSOBSTETRICS i Peer reviewed
JPOG JUL/AUG 2013 • 146
INTRODUCTION
Most women presenting with complications in early pregnancy are assessed, diagnosed
and managed at early pregnancy assessment units (EPAUs). These units aim to provide
thorough assessments, access to specialist investigations (scan, human chorionic gon-
adotrophin [hCG]), a rapid turnaround of results, and co-ordination of further manage-
ment.
The EPAU enables continuity of care, fewer admissions, and planned follow-up. It
is beneficial in the provision of open access for GPs, and ideally patients particularly
following a previous pregnancy loss. By streamlining investigations and treatment, this
system is also more cost-effective.
For women who have had previous pregnancy complication, a familiar setting and
ongoing support in a future pregnancy is a valued service.
A downside of the EPAU system is that it is often only available at limited times,
thus for complications occurring outside of these hours, patients require ward contact
numbers and more frequent inpatient based care.
HYPEREMESIS GRAVIDARUM
Over 50% of women suffer from nausea in pregnancy. Hyperemesis gravidarum is the
inability to maintain hydration, resulting in dehydration and ketonuria as a result of
nausea and vomiting in pregnancy. It affects between 0.1% and 1% of women. Patients
become dehydrated and ketonuric, develop an electrolyte imbalance (hyponatraemia and
hypokalaemia), and in severe untreated cases a nutritional (thiamine) deficiency culmi-
nates in Wernicke’s encephalopathy.
Management of Early Pregnancy Complications
Harriet Pugsley, MB ChB, MRCOG; Judith Moore, MRCOG
OBSTETRICS i Peer reviewed
JPOG_JulAug_2013_COMBINE_Final.indd 146 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 147
OBSTETRICS I Peer revIewed
Symptoms develop around 6–8 weeks of gesta-
tion and are directly related to levels of hCG, peak-
ing towards the end of the first trimester before
settling in the second trimester. Women present
with nausea, weakness, vomiting, and occasionally
ptyalism (inability to swallow saliva). On examina-
tion, there are signs of dehydration and tachycardia
with or without hypotension. Those who have suf-
fered with hyperemesis in a previous pregnancy are
more likely to develop similar symptoms in subse-
quent pregnancies. Nausea and vomiting present-
ing after 12 weeks of gestation is not hyperemesis
gravidarum.
The proposed pathophysiology behind hy-
peremesis is related to the hCG. The hCG molecule
has common alpha subunit with thyroid-stimulating
hormone and is thought to exert its effect via a
temporary physiological thyrotoxicosis; there may
be evidence of a raised free thyroxine and a low
thyroid-stimulating hormone. This accounts for the
timing of the onset and settling of symptoms in cor-
relation with levels of hCG. There may also be psy-
chological and cultural factors.
In all presentations, a multiple or molar preg-
nancy should be excluded as these conditions also
result in an increased hCG level.
Many women with mild symptoms are managed
in the community; once women are ketotic and un-
able to maintain hydration, admission is necessary.
For the majority of women, intravenous rehydration
with antiemetic medication, on a day case basis
without admission, is sufficient to break the cycle
and allow the patient to re-establish oral intake.
Many units have managed to expand the role of the
EPAU to incorporate this service and release capac-
ity on the admitting wards. However, more severe
cases do require an inpatient admission, further
rehydration, antiemetics, thiamine supplements,
daily electrolyte analysis, and a gradual resumption
of oral intake and monitoring of ketonuria. Intrave-
nous rehydration with normal saline or Hartmann’s
with potassium supplementation and monitoring of
electrolytes and ketonuria should be employed. In-
fusions containing dextrose should be avoided as
they may precipitate Wernicke’s encephalopathy.
Women requiring admission are intravascularly de-
hydrated posing an increased thrombotic risk, thus
thromboprophylaxis should be considered. Rarely,
intractable cases may require treatment with ster-
oids to relieve symptoms.
Hyperemesis is not uncommon; the majority of
cases can be successfully treated as a day admis-
sion to an EPAU. The outcome for the both preg-
nancy and patient are excellent.
MISCARRIAGE
Fifteen percent to 20% of pregnancies end in mis-
carriage.
Miscarriage was traditionally classified as
threatened, inevitable, complete, incomplete, or
missed. Most women present with pain and bleed-
ing in early pregnancy. Alternatively, miscarriage
may be diagnosed at the early dating scan with
no prior warning symptoms, this is classified as a
missed miscarriage.
Practice points I
Hyperemesis
• 50% of women suffer from nausea in pregnancy
• 0.1–1% of women suffer from hyperemesis gravidarum
• Symptoms peak at the end of the first trimester
• In severe cases, there is a risk of Wernicke’s encephalopathy re-
sulting from a nutritional imbalance of reduced thiamine
• Treatment includes intravenous rehydration, electrolyte monitor-
ing and restitution, antiemetics, thiamine supplementation and,
in non-resolving cases, steroid therapy
• Molar and multiple pregnancies should be excluded
JPOG_JulAug_2013_COMBINE_Final.indd 147 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 148
OBSTETRICS I Peer revIewed
An episode of pain or bleeding in an early
pregnancy subsequently demonstrated to be a vi-
able pregnancy is termed a threatened miscarriage.
Traditionally, classifications of complete, in-
complete and inevitable miscarriage have been
used. Diagnosis and management are based on
clinical findings: (1) examination of vaginal loss
for products of conception, and (2) examination
of the cervical, to determine whether it is open or
closed. More recently, there has been a tendency
to be increasingly conservative in management of
miscarriage with routine use of ultrasound to aid
management. Once an intrauterine pregnancy is
demonstrated by ultrasound, with uncertain viabil-
ity, a follow-up scan will be arranged 10–14 days
later to determine whether the pregnancy is ongo-
ing or not. If initial symptoms are of heavy bleeding,
or if they worsen between appointments, then med-
ical intervention in terms of a surgical evacuation
may be necessary at an earlier stage. Patients with
a pregnancy of unknown viability, who are waiting
for follow-up, must be informed of the risk of pain
and bleeding between appointments and be given
appropriate contact numbers and advised to return
if their bleeding becomes heavy.
If there are symptoms (or with minimal symp-
toms), treatment of miscarriage is based on three
alternatives: expectant, medical, or surgical. Ex-
pectant management employs awaiting the natural
course of events, for the products to pass sponta-
neously. Medical management involves combina-
tions of oral or vaginal prostaglandins to induce the
completion of miscarriage. Surgical management
involves an operation, usually vacuum aspiration,
to remove any remaining products of pregnancy.
There are risks and benefits associated with
all approaches:
1. Expectant management requires a longer fol-
low-up, often multiple visits, and is associated
with more prolonged and heavier bleeding.
Owing to the unpredictable of length of follow-
up and number of visits associated with expect-
ant management, many women veer away from
this option. There is a higher risk of treatment
failure and an increased need for later surgery.
This method of treatment may be more effec-
tive for those who are symptomatic of pregnan-
cy loss (ie, those presenting with an incomplete
miscarriage) because the process of spontane-
ous loss will have already begun. The timescale
for completion of expectant management can
extend over 2–6 weeks with the emphasis on
planned follow-up at specified intervals for
assessments usually with ultrasound scans
to ensure all products have passed. Although
figures differ depending on the length and type
of follow-up that patients have (clinical versus
ultrasound), around 80% complete without
intervention. When counselling women, the
possibility of a period of prolonged follow-up
and of the risk of incomplete loss with the
subsequent need for surgery should be stressed.
2. Medical management also aims to avoid
Practice points II
Miscarriage
• Affects up to 1 in 5 pregnancies
• Treatment options are expectant, medical, and surgical
• Expectant management aims to avoid surgery, may result in
prolonged follow-up with a risk of heavier bleeding and failed
treatment
• Medical management aims to avoid surgery, may be uncomfort-
able with heavier bleeding and risk of later surgery
• Surgical management allows early completion of treatment with
the risk of surgical and anaesthetic complications
• There is no difference in post-treatment conception rates regard-
less of method of management
JPOG_JulAug_2013_COMBINE_Final.indd 148 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 149
OBSTETRICS I Peer revIewed
surgery and is an accelerated method of
conservative management with a more predict-
able timescale of completion. It can be carried
out as an inpatient or outpatient procedure,
provided support is available for women at
home if needed. If managed in the community,
follow-up is necessary once again, to ensure
completion and that ongoing complications are
recognized and treated.
3. The benefit of surgical management is to
limit the need for prolonged follow-up,
reduce the number of symptomatic days, and
for early closure of treatment. Complications,
although infrequent, include uterine perforation,
cervical trauma, incomplete evacuation, the
risk of the anaesthetic, and a slightly higher
infection rate when compared with expectant
management.
The miscarriage treatment trial and subse-
quent Cochrane reviews (2006, 2010) have con-
cluded that there is no superior method of man-
agement and have recommended that the woman’s
preferences are taken into account when planning
care; treatment should therefore be patient-guided,
based on an informed decisions.
Importantly, the long-term follow-up of the
miscarriage treatment trial concluded that there
is no difference in later conception rates follow-
ing the different approaches to management. It has
also been suggested that empowering patients by
choice in their management reduces subsequent
anxiety and depression rates. Women who have
been involved in the decision-making process have
subsequently scored more favourably on quality-of-
life outcome questionnaires.
recurrent MiscarriageRecurrent miscarriage is defined as three or more
consecutive miscarriages and affects 1% of couples
trying to conceive.
It is recommended that investigations into
why women have miscarried should be implement-
ed following the third consecutive miscarriage.
For the majority of women, the results are normal.
Prognosis for the future worsens with increasing
numbers of pregnancy losses, advancing maternal
age, normal histopathology, and a normal parental
karyotype. Recurrent miscarriage is particularly
devastating for couples as they may be able to
conceive relatively easily but not carry an ongoing
pregnancy; they may thus need increased support
and counselling. EPAUs arrange specialist follow-
up and offer a patient educational role with the
provision of information combined with access to
support groups and counselling.
Following recurrent miscarriage, investiga-
tions may include screening for antiphospholipid
syndrome, karyotyping the products of conception,
diagnostic imaging for structural abnormalities, and
parental karyotyping looking for balanced translo-
cations. The incidence of controlled diabetes and
thyroid disease are no different in this population
when compared with that of the general public.
As stated, for the majority of patients, inves-
tigations will be normal and no cause is identified.
In a minority, there may be a treatable haemato-
logical factor such as antiphospholipid syndrome.
In the case of recurrent miscarriage, positive serum
Practice points III
Recurrent miscarriage
• Defined as three or more consecutive miscarriages
• Affects 1% of couples
• For the majority of patients, all investigations are normal
• A minority of patients will have antiphospholipid syndrome,
which can be treated with heparin and aspirin to improve the
chance of an ongoing pregnancy
JPOG_JulAug_2013_COMBINE_Final.indd 149 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 150
OBSTETRICS I Peer revIewed
blood results on two separate occasions 12 weeks
apart with a history of three consecutive pregnancy
losses before 10 weeks are necessary to confirm
antiphospholipid syndrome. (Positive bloods in as-
sociation with a single pregnancy loss above 10
weeks or a poor obstetric outcome before 34 weeks
secondary to placental disease also fulfil the crite-
ria for antiphospholipid syndrome.) For these wom-
en, treatment with aspirin and heparin will improve
the chance of an ongoing pregnancy. Without treat-
ment, the likelihood of a term pregnancy may be as
low as 10%; with treatment, the miscarriage rate
may be reduced by up to 54%.
MOLAR PREGNANCY
Molar pregnancies are rare, affecting roughly 1 in
700 conceptions. They are the result of an abnor-
mal conception and can be either complete moles
or partial moles. They are more common at the ex-
tremes of reproductive ability and among people
living in Asia. The risk of recurrence is low: 1–2%.
Complete moles are diploid, contain only pa-
ternal chromosomes, and are either the result fer-
tilization of an empty ovum with duplication of a
single sperm or dispermic fertilization. They contain
no fetal tissue.
Partial moles are triploid in nature, may con-
tain fetal parts, and are the result of dispermic fer-
tilization.
The concern with molar conceptions is the
risk of progression to neoplasia if left untreated.
Molar pregnancies can present as a miscarriage
or, more rarely, after a normal pregnancy. They are
associated with characteristic appearances on an
ultrasound scan (snow storm/placental vacuoles).
Molar pregnancies are usually detected in the
first trimester because women either present with
bleeding, resulting in an early ultrasound scan, or
are suspected on the dating ultrasound scan. They
are also associated with a raised hCG level, and the
condition should be excluded in women presenting
with hyperemesis.
Diagnosis is sometimes retrospective follow-
ing histological examination of products of concep-
tion removed following a surgical evacuation. In
cases of miscarriage, products should be screened
for gestational trophoblastic disease.
Once suspected, the majority are easily treat-
ed by surgical evacuation alone, with histological
confirmation. However, owing to the low but poten-
tially life-threatening risk of ongoing trophoblastic
tissue and neoplasia, registration and referral to
specialist tertiary care is crucial.
Due to the nature of hydatidiform gestational
trophoblastic disease/neoplasia (GTN) with the po-
tential for progressive disease, long-term follow-up
is necessary. This involves prolonged measurement
of urine hCG levels to ensure disease-free status,
avoidance of future pregnancy until hCG levels have
been normal for 6 months, and further monitoring
after all future pregnancies.
Treatment with surgical evacuation and pro-
longed monitoring of hCG levels are usually all that
is required, with repeat monitoring of levels in sub-
sequent pregnancies. All UK cases of gestational
Practice points IV
Molar pregnancy
• Affects 1 in 700 conceptions
• The concern is the risk of progression to neoplasia if unrecog-
nized or untreated
• All cases must be registered at a tertiary referral centre and re-
quire long-term follow-up
• The risk of recurrence is low (1–2%)
• Cases which have progressed to neoplasia are responsive to
treatment; 5–6% require chemotherapy with a 98% cure rate
JPOG_JulAug_2013_COMBINE_Final.indd 150 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 151
OBSTETRICS I Peer revIewed
trophoblastic disease are monitored via a national
registration programme consisting of three centres
based in Sheffield, Dundee, and Charring Cross.
hCG is a marker of ongoing trophoblastic dis-
ease. If levels are persistently elevated, further
treatment is required; this is usually medical with
methotrexate and, in rare cases, chemotherapy; a
repeat evacuation is not usually helpful. GTN is
extremely responsive to treatment. Chemotherapy
is necessary in only 5–6% of cases with a greater
than 98% cure rate and is more commonly required
following complete rather than partial molar preg-
nancies.
Following diagnosis and treatment, women
are advised to delay further conception until 6
months following normalization of the hCG levels
and a year after completion of chemotherapy. Bar-
rier contraception is advised; hormonal contracep-
tion is an option once hCG levels have normalized.
If contraception has been commenced before diag-
nosis, there may be a slight increase in the risk of
gestational trophoblastic disease progressing to
neoplasia with combined preparations, but there
is no evidence of detriment from single agent pro-
gestogens.
Thankfully, the recurrence risk is low, so wom-
en who continue to have a future pregnancy can be
reassured that 98 out of 100 conceptions will not be
a further molar pregnancy. In the women in whom a
second molar pregnancy does occur, it is usually of
the same histological type as the first.
ECTOPIC PREGNANCY
Ectopic pregnancy complicates approximately 1 in
50 pregnancies. Risk factors include previous ec-
topic pregnancy, a history of pelvic infection, or
past pelvic surgery. If not recognized or managed
inappropriately, it can become a life-threatening
condition as a result of invasion and rupture of a
blood vessel, resulting in massive pelvic haemor-
rhage; this can happen with or without rupture of
the Fallopian tube.
Presentation, as with most early pregnancy
complications, is with pain with or without vaginal
bleeding, or as collapse with associated concern-
ing symptoms of dizziness and fainting. Diagno-
sis is based on history, clinical examination, and
investigations, including transvaginal ultrasound
scan, serial hCG measurements, and laparoscopy.
In a normal pregnancy, the serum hCG level should
rise by at least 60% in 48 hours, although anything
above 30% can be consistent with normal early
pregnancy. The hCG levels in an ectopic pregnancy
may initially rise at a normal rate, but by the time
of presentation the rise is usually suboptimal. Once
suspected, ectopic pregnancies might be suggested
by ultrasound evidence of gestational tissue out-
side of the body of the uterus with an empty uter-
ine cavity. However, laparoscopy remains the gold
standard for confirmation of diagnosis and allows
concurrent treatment.
Although diagnosis can be suggested by pel-
vic ultrasound and hCG, confirmation can often
take several days because of presenting features
in common with other early pregnancy complica-
tions, uncertainty on ultrasound scan, and the need
Practice points V
Ectopic pregnancy
• Affects 1 in 50 pregnancies
• The main concern is the risk of life-threatening bleeding
• Management options depend largely on presentation and include
mainly surgery or methotrexate (rarely, expectant management
may be employed in the case of resolving trophoblast)
• Salpingotomy is only recommended during surgical treatment if
there is a concern that the contralateral tube is non-functional
JPOG_JulAug_2013_COMBINE_Final.indd 151 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 152
OBSTETRICS I Peer revIewed
for serial blood levels. In the stable patient in this
situation, the 48-hour hCG trend and symptoms are
crucial factors in determining timing and options
for treatment. In cases of haemodynamic insta-
bility secondary to a suspected bleeding ectopic
pregnancy, emergency surgery is necessary either
by laparotomy or laparoscopy, in order to stem the
bleeding and remove the ectopic as rapidly and
safely as possible.
Once diagnosed, there are several options for
the management of an ectopic pregnancy, depend-
ing mainly on presentation and individual circum-
stances. For example, in the acute emergency due
to a bleeding ectopic, urgent surgery is required
as a lifesaving procedure. However, the majority
of ectopic pregnancies are treated by scheduled
laparoscopic surgery or by medical management
with methotrexate. Conservative management is
also an option with resolving trophoblast (the trend
is for the hCG level to fall). Laparoscopic surgery
may be initially diagnostic to confirm the ectopic
pregnancy as the cause of pain and then therapeu-
tic by salpingectomy or salpingotomy as definitive
treatment.
Any surgery performed will have implications
for future fertility potential. At the time of surgery,
both Fallopian tubes should be assessed as the
health of the unaffected Fallopian tube is also rele-
vant. The surgical recommendation is for salpingec-
tomy to be performed if the contralateral tube is
deemed healthy and that salpingotomy should only
be performed if there is doubt about fertility with
the remaining tube. This is because preservation of
the Fallopian tube following an ectopic pregnancy
retains the risk of a subsequent ectopic pregnancy
in the same tube, whereas a salpingectomy removes
this risk. The fertility rate from a single functional
tube should be sufficient to allow conception.
If there is a diseased contralateral Fallopian
tube (such as a hydrosalpinx) and a salpingectomy
is performed, future options for conception are like-
ly to depend on in vitro fertilization. The limitation
of visual assessment of the tube should be remem-
bered as it will give an indication of gross disease
but not assess tubal patency for which a dye test is
more accurate.
Owing to the risk of life-threatening bleeding
from an ectopic pregnancy and the need to be ab-
solutely certain that a viable intrauterine pregnancy
has been excluded, there are rigid criteria to be met
before medical management becomes an option.
It is recommended that ultrasound findings
should show a suspected ectopic mass less than 3
cm in size with minimal free fluid and no visible
fetal heart. There should be minimal pain on ex-
amination and a suboptimal rise in hCG which is
initially less than 3,000 IU/L. (The Royal College of
Obstetricians and Gynaecologists disseminated na-
tional recommendation, from which individual units
For the majority of patients with recurrent miscarriage, investigations are normal and no cause is identified.
JPOG_JulAug_2013_COMBINE_Final.indd 152 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 153
OBSTETRICS I Peer revIewed
set their own treatment pathways). These criteria
are employed to enable patient selection so that
medical management is an option for appropriate
patients who are unlikely to have spontaneous
bleeding whilst undergoing treatment. Success of
treatment is assessed by subsequent hCG estima-
tions.
The main benefit of methotrexate is the avoid-
ance of surgery. The downside includes the low risk
of failure of treatment in cases in which ongoing
trophoblast results in bleeding from the ectopic
site and the need for surgical intervention.
Methotrexate has antifolate properties and is
teratogenic, so patients must be suitably advised
to delay conception following treatment, and care
must be taken to encourage folic acid prior to and
during subsequent conceptions.
Expectant management can be employed for
cases of resolving trophoblast, which may either be
a resolving ectopic pregnancy or an early pregnan-
cy miscarriage. For safety, levels of hCG must be
decreasing and women must be able to seek appro-
priate help and attend hospital easily should they
need to. Follow-up needs to be rigorous to ensure
that the hCG titres have returned to non-pregnant
levels and complications are not missed.
For women who have experienced an ectopic
pregnancy, fears for the future include the risk of
a recurrent ectopic pregnancy, concerns regarding
the ability to conceive again, and fear relating to
the health risks of a recurrent ectopic pregnancy.
It may be particularly worrying for women to plan a
further pregnancy if they have already experienced
life-threatening emergency surgery from a ruptured
ectopic or a previous pregnancy complication. If one
Fallopian tube is diseased, it is likely that the other
tube is also affected, and spontaneous conception
may be delayed or impossible. Therefore fertility
may well depend on in vitro fertilization which may
not be an option for many women (limited National
Health Service funding, personal finances, and
emotional reasons).
CONCLUSION
For the majority of women, conception, pregnancy,
and birth are straightforward, resulting in an uncom-
plicated confinement, delivery, and healthy baby.
For women who do experience complications,
explanation, advice, follow-up, and necessary in-
vestigations will help when planning for the future
and deciding whether to try for a further pregnancy.
Women who are informed of the long-term effects of
treatment, for example, that surgical/medical man-
agement following a miscarriage does not convey
differences in conception rates, and who are given
an informed choice regarding their management,
may find it easier to come to terms with their loss
and have less anxiety and depression long-term. It
must be remembered that each case is individual,
all circumstances are different and women need
Women who are informed
of the long-term effects
of treatment... and who are
given an informed choice
regarding their management,
may find it easier to come
to terms with their loss
and have less anxiety and
depression long-term.
JPOG_JulAug_2013_COMBINE_Final.indd 153 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 154
OBSTETRICS I Peer revIewed
FURTHER READING
Royal College of Obstetricians and Gynaecologists. The manage-ment of early pregnancy loss. Green-top Guideline No. 25. London: RCOG, 2006.
Trinder J, Brocklehurst P, Porter R, Read M, Vyas S, Smith L. Management of miscarriage: expectant, medical or surgical? Results of randomized controlled trial (miscarriage treatment (MIST) trial). BMJ 2006;332:1235.
Smith LFP, Ewings PD, Quinain C. Incidence of pregnancy after expectant, medical or surgical management of spontaneous first trimester miscarriage. Long term follow-up of miscar-riage treatment (MIST) randomized controlled trial. BMJ 2009;339:b3827.
Nanda K, Peloggia A, Grimes DA, Lopez LM, Nanda G. Expect-ant care versus surgical treatment for miscarriage. Cochrane Database Syst Rev 2006. Issue 2. Art. No.: CD003518. pub. 2.
Neilson JP, Gyte GML, Hickey M, Vaquez JL, Don L. Medical treatments for incomplete miscarriage (less than 24 weeks). Cochrane Database Syst Rev 2010. Issue 1. Art. No.: CD007223. pub. 2.
Neilson JP, Hiskey M, Vaquez JL. Medical treatment for early fetal death (less than 24 weeks). Cochrane Database Syst Rev 2006. Issue 3. Art. No.: CD002253. pub. 3.
Rai RS, Clifford K, Cohen H, Regan L. High prospective fetal loss rate in untreated pregnancies of women with recurrent miscarriage and antiphospholipid antibodies. Hum Reprod 1995;10:3301–3304.
Royal College of Obstetricians and Gynaecologists. The manage-ment of recurrrent miscarriage. Green-top Guideline No. 17. London: RCOG, 2011.
Royal College of Obstetricians and Gynaecologists. The manage-ment of gestational trophoblastic neoplasia. Green-top Guide-line No. 38. London: RCOG, 2010.
Royal College of Obstetricians and Gynaecologists. The manage-ment of tubal pregnancy. Green-top Guideline No. 21. London: RCOG, 2004.
© 2012 Elsevier Ltd. Initially published in Obstetrics, Gynaecol-ogy and Reproductive Medicine 2012;22(3):76–80.
About the AuthorsHarriet Pugsley is Specialist Registrar in Obstetrics and Gynae-cology. Judith Moore is Consultant Obstetrician and Gynaecolo-gist at Nottingham United Hospitals NHS Trust, Nottingham City Hospital, Nottingham, UK.
to be informed of the risks and benefits of all the
available treatment options.
The role of the EPAU, with open access and
support in future pregnancies in terms of regular
scanning and as a point of contact for the individual,
has been shown to improve the successful outcome
of future pregnancies following miscarriage. Open
access and choice in care options have also been
advocated as helping women to feel more in con-
trol of their treatment and reducing post-treatment
rates of anxiety and depression.
Besides the feeling of loss, it is normal for
women to feel guilt, blame, and anxiety. Women
will have concern for their own health, particularly
if they have required emergency treatment. Part-
ners may also harbour feelings of guilt, responsi-
bility, and helplessness for the situation, and may
have conflicting emotions with concerns regarding
the risk to their loved ones’ health in a future preg-
nancy, balanced with their desire to increase their
family.
It is important for patients who have had an
ectopic pregnancy to understand that they require
an early ultrasound in subsequent pregnancies to
exclude a recurrence and to seek a medical assess-
ment early in pregnancy.
Previously, in future conceptions, early open
access and regular specialist follow-up have been
available, enabling reassurance, investigations,
early detection of further complications, along with
support throughout subsequent pregnancies. Unfor-
tunately as healthcare services are put under ever
increasing pressure to reduce costs, open access
may no longer be deemed a financially viable ser-
vice.
It is important that appropriate investigations
are arranged and results collated and explained. It
is also necessary that patients are given appropri-
ate counselling, information, and links to support
groups should they require additional care.
JPOG_JulAug_2013_COMBINE_Final.indd 154 8/5/13 2:18 PM
Autism Spectrum Disorders Patricia Howlin, BA MSc PhD FBSP, Professor of Clinical Child Psychology
GYNAECOLOGY I Peer revIewed
JPOG JUL/AUG 2013 • 155
GYNAECOLOGY i Peer reviewed
INTRODUCTION
The lifetime prevalence of ovarian cancer in the developed world is 1–2%. It is often
described as a silent killer; however, the majority of women frequently experience symp-
toms in the months leading to diagnosis. The majority of ovarian cancers are diagnosed
at an advanced stage. In England and Wales, ovarian cancer kills more women than all
of the other gynaecological malignancies combined.
Pregnancy, breastfeeding, and use of the oral contraceptive pill all appear to pro-
tect against the development of epithelial ovarian cancer; the lower incidence seen in
less developed countries may be related to a higher birth rate.
Patients with ovarian cancer are best managed by multidisciplinary teams. These
usually include nurse specialists, medical oncologists, histopathologists, radiologists,
palliative care specialists, and gynaecological oncologists, in collaboration with the
patients and their families.
TYPES OF OVARIAN CANCER
Primary ovarian tumour types include epithelial, sex cord–stromal and germ cell tu-
mours. Tumours not specific to the ovaries also occur, such as sarcomas and lymphomas.
Metastatic tumours from breast, stomach, and endometrial primaries are not uncommon.
epithelial TumoursMore than 80% of ovarian cancer is epithelial in origin. The most common subtype
is serous, accounting for about 50%, followed by endometrioid, mucinous, clear cell,
transitional (Brenner), mixed, and undifferentiated tumours. These have previously been
Ovarian Cancer: Current Management and Future Directions
Siân E Taylor, BSc, MB ChB, MRCOG, MD; John M Kirwan, MB ChB, MRCOG
JPOG_JulAug_2013_COMBINE_Final.indd 155 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 156
GYNAECOLOGYGYNAECOLOGY I Peer revIewed
treated as essentially the same disease with dif-
fering histology; however, it is becoming increas-
ingly evident that they behave as distinct entities.
For example, endometrioid and clear cell cancers
are strongly linked with endometriosis, whilst many
mucinous cancers originate in the appendix and evi-
dence increasingly points to serous tumours arising
from dysplastic endometrium in the distal fallopian
tube. Response to chemotherapy varies; serous tu-
mours tend to be highly chemosensitive, but clear
cell and mucinous tumours are more resistant to
conventional chemotherapy.
Primary peritoneal cancer is histologically
indistinguishable from metastatic serous ovarian
cancer. It is diagnosed in the absence of any clear
ovarian primary. Treatment is the same as for ovar-
ian cancer, although, as there is often no mass to
debulk, chemotherapy is more often used as the
primary treatment.
Borderline ovarian tumours are not truly can-
cers but are termed ‘borderline’ because they show
histological features that are intermediate between
benign and malignant tumours. They are staged in
exactly the same way and sometimes spread be-
yond the ovary to produce non-invasive implants in
the omentum and the peritoneum. They can recur
after long periods; cases have been documented
with disease returning over 30 years after the initial
presentation. They are typically found in a younger
population compared with epithelial cancers, one-
third occur in women under the age of 40. The main
treatment is surgical excision, and opinions differ
in the extent of surgery required. It is probable that
they represent premalignant disease for low-grade
ovarian carcinomas.
Sex Cord–Stromal TumoursSex cord–stromal tumours account for approximate-
ly 7% of all malignant ovarian tumours. They arise
from a combination of the hormone-producing cells
of the ovary and stromal fibroblasts. Seventy per-
cent of all malignant sex cord–stromal tumours are
granulosa cell tumours. The majority occur in wom-
en in their sixth decade, although a small proportion
arises in young women and prepubertal girls. Gran-
ulosa cell tumours may secrete sex hormones; most
secrete oestrogen (although androgen-secreting va-
rieties do occur) potentially leading to endometrial
hyperplasia and carcinoma. Presenting symptoms
include abdominal distension, acute abdominal
pain, and abnormal vaginal bleeding. As most pre-
sent at an early stage, the prognosis is good. Treat-
The histological subtypes of epithelial ovarian cancer are distinct entities, requiring different treatment strategies.
JPOG_JulAug_2013_COMBINE_Final.indd 156 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 157
GYNAECOLOGYGYNAECOLOGY I Peer revIewed
ment is principally surgical with platinum-based
chemotherapy for advanced disease. Surgical treat-
ment is as for epithelial ovarian cancer, although in
young women with early disease, fertility preserva-
tion is an option. Other stromal tumours are rare
and include thecomas, fibromas, Sertoli-Leydig cell
tumours, and gynandroblastomas.
Malignant Germ Cell TumoursMalignant germ cell tumours occur chiefly in girls
and young women. The most common variety is the
dysgerminoma, the counterpart to the seminoma in
the male. Other types include the yolk sack tumour,
embryonal carcinoma, polyembryoma, non-gesta-
tional choriocarcinoma, and teratoma. They usually
present with abdominal pain, which is sometimes
acute, and a palpable pelvic mass. Treatment for
early-stage disease is surgical. As over 60% are
confined to one ovary at diagnosis, fertility-sparing
surgery is usual with unilateral salpingo-oophorec-
tomy or even ovarian cystectomy in selected cases
with otherwise normal ovaries. Dysgerminomas are
highly radiosensitive, but platinum-based chemo-
therapy is currently the preferred option as radio-
therapy usually results in premature ovarian failure.
Non-dysgerminoma tumours are treated with the
chemotherapy combination of bleomycin, etopo-
side, and cisplatin. Response rates are excellent
with cure rates approaching 100% in early-stage
disease and up to 75% in advanced disease.
FAMILIAL OVARIAN CANCER
Approximately 5–10% of all ovarian cancer is as-
sociated with a genetic predisposition. Individuals
carrying these gene defects have a significantly
higher risk of developing ovarian cancer than the
general population (Table 1).
A strong family history of breast and/or ovar-
ian cancer, especially at a relatively young age,
may indicate the presence of one of several pos-
sible BRCA1 or BRCA2 gene mutations. The gene
products are involved in DNA repair. These gene
mutations are particularly common amongst the
Ashkenazi Jewish population. Men carrying these
genes are at increased risk of pancreatic cancer
as well as male breast cancer. Endometrial, colon,
ovarian, and other cancers cluster in families with
the Lynch II or hereditary non-polyposis colorectal
cancer (HNPCC) syndrome. This is caused by a vari-
ety of defects in the DNA mismatch repair system.
BRCA1/2 and HNPCC are inherited in an autosomal
dominant fashion; further mutation or epigenetic
silencing of the second allele results in malignancy.
Prophylactic risk-reducing surgery is recom-
mended for BRCA1 and BRCA2 carriers when they
reach the age of 35 or have completed their fami-
lies. This usually consists of a laparoscopic bilateral
salpingo-oophorectomy. The risk of ovarian cancer
under 35 years in BRCA patients is low, and remov-
al of the ovaries has the dual advantage of reducing
the risks of both ovarian and breast cancer. HNPCC
carriers are offered hysterectomy and bilateral sal-
pingo-oophorectomy when their family is complete.
Until patients are ready to undergo surgery, annual
screening is commonly offered with cancer antigen
125 (CA 125) and transvaginal ultrasound; however,
Table 1. Approximate lifetime percentage risk of developing ovarian cancer
General populationOne first-degree relative affected under 55 yearsOne first-degree relative affected over 55 yearsTwo first-degree relatives affectedBRCA1 carrierBRCA2 carrierHNPCC carrier
1.6%5.2%3.4% 7%
28–44%27%12%
HNPCC = hereditary non-polyposis colorectal cancer.
JPOG_JulAug_2013_COMBINE_Final.indd 157 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 158
GYNAECOLOGYGYNAECOLOGY I Peer revIewed
there is no evidence that this is effective. Even if
the ovaries are removed, there is a small continuing
risk of developing primary peritoneal cancer.
The lack of one of these specific conditions
does not exclude an increased cancer risk, as not all
inherited ovarian cancer syndromes have had their
genetic origin fully characterized.
BIOMARKERS AND SCREENING
The most commonly used marker for ovarian cancer
is CA 125. This is a glycoprotein that is released
into the bloodstream by any condition that disturbs
the peritoneum, including any peritoneal cancer,
cirrhosis, congestive cardiac failure, endometrio-
sis, and pelvic inflammatory disease. Pregnancy
also causes a variable increase in serum levels. It
is therefore notoriously non-specific at low levels.
However, serous ovarian cancer can cause dramatic
increases in CA 125. This is only likely to occur
when the disease has already spread beyond the
ovary. Other tumour subtypes, especially mucinous,
often produce a more modest elevation.
Other markers are also used. Germ cell ovarian
cancers often secrete highly specific tumour mark-
ers which are useful in diagnosis and monitoring,
eg, α-fetoprotein, β-hCG, and lactate dehydroge-
nase. These should be tested, in addition to CA
125, in women under the age of 40 years with a
suspicious pelvic mass. In addition, inhibin may be
of use as a marker for mucinous and granulosa cell
tumours.
Novel markers are currently under development
both as panels and individually. Human epididymis
protein 4 has been suggested as a biomarker with
equal or greater sensitivity and specificity than CA
125, however clinical trials are awaited.
The PLCO (prostate, lung, colon, ovary) trial,
a huge US-based randomized controlled trial of
cancer screening, has recently reported. It con-
Screening the anxious low-risk patient for ovarian cancer may not be beneficial because of the risk of false-positive result.
JPOG_JulAug_2013_COMBINE_Final.indd 158 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 159
GYNAECOLOGYGYNAECOLOGY I Peer revIewed
cluded that screening the general population with
annual CA 125 and transvaginal ultrasound does
not reduce ovarian cancer mortality. Indeed, a sig-
nificant number of women had major complications
from surgery performed because of a false-positive
screening test.
There are also two large UK-based multicentre
trials investigating ovarian cancer screening. The
first is in postmenopausal women without a signifi-
cant family history of ovarian cancer and also uses a
combination of transvaginal ultrasound and CA 125
(UK Collaborative Trial of Ovarian Cancer Screen-
ing [UKCTOCS]). The second is in women with a
significant family history and is testing a panel of
biomarkers in addition to CA 125 and transvaginal
ultrasound (UK Familial Ovarian Cancer Screening
Study). Both have now finished recruitment and are
due to report in the next few years.
Screening with a combination of CA 125, ul-
trasound, and pelvic examination is commonly per-
formed for anxious patients who desire screening
for ovarian cancer. If the findings of the PLCO trial
are duplicated in the UKCTOCS trial, this practice is
likely to become hard to defend in low-risk women.
INVESTIGATION
The first symptoms of ovarian cancer usually
emerge some time before diagnosis. These com-
monly include early satiety, changes in bowel
habit, bloating, urinary frequency, and pelvic and
abdominal pain. Patients with advanced cancer of-
ten complain of abdominal swelling and discomfort
due to ascites with or without a large abdomin-
opelvic mass. Eating is often difficult, and patients
may notice weight loss, apart from the distended
abdomen. It is not uncommon for patients to pre-
sent with a swollen leg secondary to a deep vein
thrombosis. However, most small ovarian cancers
are asymptomatic when confined to the ovaries and
therefore difficult to detect.
Recent National Institute for Clinical Excel-
lence (NICE) guideline recommend that women,
especially those over 50 years old, who experience
symptoms persistent or frequent symptoms as de-
scribed above, or new-onset symptoms suggestive
of irritable bowel syndrome, should have CA 125
Table 2. FIGO staging of ovarian cancer
Stage Description
I Confined to ovaries
Ia One ovary, no ascites present containing malignant cells, no tumour on external surface, capsule intact
Ib Both ovaries, no ascites present containing malignant cells, no tumour on external surfaces, capsule intact
Ic Tumour limited to one or both ovaries with any of the following: tumour on the surface on one or both ovaries, capsule ruptured, ascites present with malig-nant cells or positive peritoneal washings
II Growth involving one or both ovaries with pelvic exten-sion
IIa Extension and/or metastases to uterus and/or fallopian tubes
IIb Extension to other pelvic tissues
IIc Tumour stage IIa or IIb but with tumour on surface of one or both ovaries, capsule ruptured, ascites pres-ent containing malignant cells or positive peritoneal washings
III Tumour involving one or both ovaries with microscopi-cally confirmed peritoneal implants outside the pelvis and/or regional lymph node metastasis
IIIa Microscopic peritoneal metastasis beyond the pelvis
IIIb Macroscopic peritoneal metastasis beyond the pelvis, 2 cm or less in greatest dimension
IIIc Abdominal implants greater than 2 cm in diameter and/or regional lymph nodes metastasis
IV Distant metastasis beyond the peritoneal cavity. Includes liver parenchymal metastasis and/or pleural effusion with positive cytology
FIGO = International Federation of Gynecology and Obstetrics.
JPOG_JulAug_2013_COMBINE_Final.indd 159 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 160
GYNAECOLOGYGYNAECOLOGY I Peer revIewed
measured. If the level is > 35, an ultrasound of the
abdomen and pelvis should be performed.
The ultrasound and CA 125 together are used
to calculate the risk of malignancy index (RMI): RMI
= U × M × CA 125, where U = ultrasound score (1
point for each of multilocular cysts, solid areas,
ascites, bilateral lesions, metastases. U = 0 for 0
points, U = 1 for 1 point, U = 3 for 2–5 points); M =
menopausal status (premenopausal = 1, postmeno-
pausal = 3); CA 125 = serum CA 125 level.
A score of 250 has been chosen by NICE to
guide triage to either surgery in a cancer centre un-
der the care of a specialist multidisciplinary team,
including subspecialist gynaecological oncologists
(≥ 250), or to care under a general gynaecologist
with an interest in gynaecological oncology in a
cancer unit (< 250).
If ovarian cancer is suspected, a computed to-
mography of the abdomen and pelvis (and thorax if
clinically indicated) should be performed, prior to
surgery, to assess the extent of the disease.
STAGING
Staging is performed to guide treatment and to
provide information on prognosis. Traditionally, this
has been achieved by performing a staging lapa-
rotomy. Information can also be gleaned from ra-
diological investigations, guided biopsy, and cytol-
ogy of ascitic or pleural fluid. The current staging
system was devised by International Federation of
Gynecology and Obstetrics (FIGO) (see Table 2).
TREATMENT
Treatment in ovarian cancer depends upon the stage
at presentation and the histological subtype. Non-
epithelial cancers are discussed earlier. In general,
epithelial ovarian cancers are treated with a combi-
nation of surgery and chemotherapy. Except in very
early disease, treatment is rarely curative but it can
provide symptom relief and prolong life.
Traditionally, ovarian cancer is treated with a
staging and debulking laparotomy followed by six
cycles of chemotherapy. Second-look laparotomy,
where a second surgical debulking procedure is
performed after completion of chemotherapy, is not
beneficial. Data from the EORTC 55971 trial, which
compared interval debulking surgery performed
midway between six cycles of chemotherapy with
the traditional laparotomy and six cycles of post-
operative chemotherapy, suggested that neoadju-
vant chemotherapy in bulky stage IIIc/IV disease
did not adversely affect prognosis and that interval
debulking is associated with a lower post-operative
morbidity and mortality. The CHORUS (CHemother-
apy OR Upfront Surgery) trial also addresses this
question and its publication is awaited.
In light of this, interval debulking surgery, per-
Advanced epithelial ovarian cancer is difficult to cure.
JPOG_JulAug_2013_COMBINE_Final.indd 160 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 161
GYNAECOLOGYGYNAECOLOGY I Peer revIewed
formed midway through chemotherapy, is gaining
popularity.
SURGERY
Surgical treatment, whether primary or as an inter-
val debulking procedure, involves a midline laparot-
omy, sampling of ascitic fluid or peritoneal wash-
ings for cytology, full assessment by palpation of
all peritoneal surfaces and biopsy of any suspicious
areas, removal intact of any encapsulated masses
or debulking of tumour, sampling of suspicious
pelvic and para-aortic lymph nodes, omentectomy,
total abdominal hysterectomy, and bilateral salpin-
go-oophorectomy. The aim is to completely remove
all visible disease. This is thought to promote an
optimum response to chemotherapy. Surgery also
provides ample material for diagnostic histological
assessment. The evidence for debulking improving
chemosensitivity is not absolute. Many argue that
the ability to perform optimal cytoreduction is a
more a reflection of favourable tumour biology with
an intrinsically better prognosis than the surgery
itself influencing outcome. However, it has been
demonstrated that survival in stage III disease is
improved by the primary surgery being performed by
a specialist gynaecological oncologist, rather than
an obstetrician/gynaecologist or a general surgeon.
Surgical aggressiveness varies considerably
between continents, countries, and individual units.
Some studies have shown improvements in survival
with radical surgery, including liver resection, bowel
resection, and splenectomy. Published case series
show small or no increases in mortality and morbid-
ity. However, these studies usually involve highly se-
lected cases from single institutions and may not be
directly applicable to wider practice. Some disease
remains unresectable to even the most adventurous
surgeon; this includes small bowel mesenteric dis-
ease and disease of the portal triad.
Recent NICE guidance recommends that lym-
phadenectomy is restricted to removal of clinically
involved nodes. Full pelvic and para-aortic node dis-
section in suspected stage I disease is not recom-
mended.
Less aggressive surgery is preferred in early
epithelial ovarian cancer in young women who wish
to preserve their fertility. About 8% of stage I epi-
thelial ovarian cancers occur in women under the
age of 35. A proportion of these will not have com-
pleted childbearing and may wish to consider fertil-
ity-sparing surgery. Suitable patients include those
with stage IA, grade 1 or possibly grade 2 disease.
Such conservative surgery would typically consist
of peritoneal fluid cytology, unilateral salpingo-
oophorectomy, omental biopsy, and careful inspec-
tion of the contralateral ovary and nodal chain. One
case series describes 282 women treated conserva-
tively for epithelial ovarian cancer. Just over 30%
The decision to adopt palliation in patients with ovarian cancer can be challenging.
The decision to adopt palliation in patients with ovarian cancer can be challenging.
JPOG_JulAug_2013_COMBINE_Final.indd 161 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 162
GYNAECOLOGYGYNAECOLOGY I Peer revIewed
subsequently went on to have term deliveries. Four
percent died of conditions related to their disease.
Surgery also plays a role in palliative care of
patients with ovarian cancer. Bowel obstruction is
common in the end stages of the disease; post-mor-
tem studies of women dying with ovarian cancer
revealed bowel obstruction in almost 50%. Before
contemplating such surgery, consideration must be
given as to whether it is appropriate. Surgery should
only be performed if it has a reasonable chance of
success, and risks need to be carefully balanced
against potential symptom relief. Contraindications
to surgery include patient refusal, rapidly accumu-
lating acsites, high obstruction, multiple levels of
obstruction, and poor nutritional status. Decisions
involving palliative surgery should involve the mul-
tidisciplinary team and careful discussion with the
patient and her relatives.
CHEMOTHERAPY
Patients should have histological confirmation of
their ovarian cancer prior to starting chemotherapy.
This may be obtained through image-guided percu-
taneous biopsy, or where this is not possible or the
results are inadequate, by laparoscopic biopsy.
The current standard first-line chemotherapy
regimen for ovarian cancer involves intravenous ad-
ministration of a platinum-based drug with a taxa-
ne, usually paclitaxel, given 3 weekly for six cycles.
Most units prefer carboplatin to cisplatin as it has
a less toxic side effect profile. Evidence for the use
of paclitaxel is drawn from its efficacy in relapsed
ovarian cancer. However, paclitaxel significantly in-
creases the risk of neuropathy when compared with
carboplatin alone, and some patients have anaphy-
lactic reactions to taxanes. For these reasons, it
is not universally used. Chemotherapy can also be
given via the intraperitoneal route; however, cur-
rently in the UK its use is not recommended outside
of clinical trials.
Epithelial ovarian cancer is one of the more
chemosensitive solid tumours, and complete clini-
cal and radiological response occurs in up to 50%
with the above regimen. Conversely, 20–30% will
show no evidence of response. Unfortunately, the
majority of patients with advanced ovarian cancer
will relapse. Chemotherapy for recurrent disease is
determined in part by the length of time before re-
lapse occurs. If it is more than 6 months afterwards,
it is potentially platinum-sensitive and, unless con-
traindicated, a regimen containing platinum will be
used again. A taxane is likely to be included, espe-
cially if not used initially. Paclitaxel is sometimes
used at a lower dose at more frequent intervals;
this appears to reduce the adverse effects experi-
enced. Response rates are in the order of 30%. If
relapse occurs within 6 months, second-line drugs,
such as liposomal doxorubicin and topotecan, may
be considered. As response rates are low, approxi-
mately 10–20%, the choice of drug is made bearing
in mind side effect profiles and ease of administra-
tion.
Novel chemotherapeutic agents are constantly
being developed, some of the most promising effect
the functioning of vascular endothelial growth fac-
Practice points
• Ovarian cancer is best managed within a cancer centre by a multidisciplinary team
• Women with symptoms suggestive of ovarian cancer should have a CA 125 performed as an initial screen, followed by ultrasound if abnormal
• Ovarian cancer screening with ultrasound and CA 125 does not appear to be helpful in the low-risk population
• Treatment involves a combination of surgery and platinum-based chemotherapy
• Fertility-sparing surgery is possible in women with early-stage disease
JPOG_JulAug_2013_COMBINE_Final.indd 162 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 163
GYNAECOLOGYGYNAECOLOGY I Peer revIewed
tor. Interim analysis of the ICON 7 trial of standard
therapy with or without bevacizumab (a monoclo-
nal antibody to vascular endothelial growth factor)
shows a sustained improvement in progression-free
survival in a subgroup of women with advanced dis-
ease and suboptimal surgical debulking.
Hormonal therapies are occasionally used;
these probably act by reducing oestrogen activity
and include tamoxifen, aromatase inhibitors, and
gonadotropin-releasing hormone analogues. Re-
sponse rates of 10–15% have been achieved in re-
lapsed disease. Their main advantage is their mini-
mal side effects when compared with conventional
chemotherapy.
PALLIATIVE CARE
Palliation is an integral part of the care of pa-
tients with ovarian cancer. Maintaining a balance
between optimism and pragmatism, together with
knowing when the emphasis of care should tilt to-
wards palliation, are some of the most challenging
aspects of caring for this group of patients.
Almost all patients with advanced ovarian
cancer, whether in the terminal phase or not, will
have distressing symptoms that require treatment.
These symptoms may be due to the disease itself
or secondary to their treatment. Common symptoms
include nausea, pain, loss of appetite, constipation,
and abdominal distension. This is too small a space
to describe all the possible treatment strategies
involved, but it is worth emphasizing a few princi-
ples. It is important to take a history, examine the
patient, and review the drug chart and case notes.
Then consider what investigations may be helpful
and what drug or intervention will best treat the
most likely cause of her symptoms. These patients
are usually complex with multiple problems; how-
ever, a logical and systematic approach will help to
identify the best treatment. Subcutaneous infusions
delivered by syringe driver are useful as they permit
a steady concentration of drug and oral medications
may be poorly absorbed. Several hospices publish
their guidelines for managing symptoms in pallia-
tive care on the internet, which can be very helpful.
Finally, the physical needs of a seriously ill pa-
tient are only one facet of their care. Social, spiritu-
al, and emotional needs also need to be addressed,
both for the patient and their relatives. Specialist
oncology nurses and palliative care input is essen-
tial, as are discussions on resuscitation status and
preferred place of death. These aspects of care are
easily overlooked but can make the difference be-
tween a peaceful and a difficult death.
FURTHER READING
Agarwal R, Linch M, Kaye SB. Novel therapeutic agents in ovarian cancer. EJSO 2006;32:875–886.
Aletti GD, Gallenberg MM, Cliby WA, Jatoi AJ, Hartmann LC. Current management strategies for ovarian cancer. Mayo Clin Proc 2007;82:751–770.
Blagden S, Gabra H. Future directions in the management of epithelial ovarian cancer. Future Oncol 2008;4:403–411.
Cannistra SA. Cancer of the ovary. N Engl J Med 2004;351:2519–2529.Colombo N, Parma G, Zanagnolo V, Insinga A. Management of ovarian
stromal cell tumours. J Clin Oncol 2007;25:2944–2951.Gershenson DM. Management of ovarian germ cell tumours. J Clin
Oncol 2007;25:2938–2943.Goff BA, Mandel L, Muntz HG, Melancon CH. Ovarian carcinoma
diagnosis, results of a national ovarian cancer survey. Cancer 2000;89:2068–2075.
Markman M. New, expanded, and modified use of approved antineo-plastic agents in ovarian cancer. The Oncologist 2007;12:186–190.
NICE clinical guideline 122, Ovarian Cancer.Nickles Fader A, Rose PG. Role of surgery in ovarian carcinoma. J Clin
Oncol 2007;25:2873–2883.Rao G, Crispens M, Rothenberg ML. Intraperitoneal chemotherapy
for ovarian cancer: overview and perspective. J Clin Oncol 2007;25:2867–2872.
© 2011 Elsevier Ltd. Initially published in Obstetrics, Gynaecology
and Reproductive Medicine 2011;22(2):34–37.
About the AuthorsSiân E Taylor is Subspecialty Trainee in the Department of Gynae-cologial Oncology at Liverpool Women’s Hospital, Liverpool, UK. John M Kirwan is Consultant Gynaecological Oncologist at Liverpool Women’s Hospital, Liverpool, UK.
JPOG_JulAug_2013_COMBINE_Final.indd 163 8/5/13 2:18 PM
MIMS.com now comes with a brand new look and mobile app!Along with a redesign that helps enhance ease of access to crucial clinical information, MIMS.com can now be accessed through a FREE mobile application, available on the iTunes store.
With MIMS available across multiple platforms and devices, you can now easily and conveniently find the most up-to-date and relevant drug information anytime, anywhere.
CONNECT WITH MIMS:
www.mims.com MIMS mobile/tablet app facebook.com/mimscom
Journal of Paediatrics, Obstetrics & Gynaecology
MIMScomNew206x276_0613.indd 1 6/4/13 3:45 PM
Imaging Paediatric Brain Tumours
Tang Phua Hwee, MBBS, FRCR, MMed Diagnostic Radiology
PAEDIATRICS i Peer reviewed
JPOG JUL/AUG 2013 • 164
Constipation in Infants and Children
Taya Dowling, BMedSci, MB BS; Scott Nightingale, BMed(Hons), MClinEpid, FRACP
REMEMBER
• Constipation is a common paediatric presentation.
• Constipation is defined based on the frequency of stooling (which varies widely
depending on the age of the child), and more importantly the consistency, size and
difficulty with which stools are passed (see the box on page 165).1
• Constipation arising beyond the neonatal period is most often functional (ie, does not
result from any identifiable organic pathology), and may be perpetuated by voluntary
withholding of stool to avoid painful defaecation.
• There are several ‘red flags’ that should prompt further investigation for a contrib-
uting medical or surgical condition, but these are uncommon (see the box on page
165).2,3
• Faecal incontinence is thought to be related to over distension of the rectum with
stool, shortening of the anal canal and resulting impairment of normal continence
mechanisms. Incontinence is not deliberate or caused by laziness on the part of the
child.
• Untreated constipation and faecal incontinence (encopresis) can have a significant
psychosocial impact on a child.
• Management of constipation is often a long- term process that requires the com-
plementary approaches of careful education of the child and parents, behavioural
techniques, laxative agents and review.4
ASSESSMENT
• The focus should be on identifying the rare child with an organic cause for consti-
PAEDIATRICS i Peer reviewed
JPOG_JulAug_2013_COMBINE_Final.indd 164 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 165
PAEDIATRICSPAEDIATRICS I Peer revIewed
pation, and determining whether the child has
faecal impaction.
• The history should include a detailed descrip-
tion of the child’s stool, stool frequency, incon-
tinence, withholding behaviour and any symp-
toms associated with defaecation, such as pain,
bleeding and straining.
• Important aspects of the history include age at
onset, growth trends, diet history and the pres-
ence of ‘red flags’ (see the box on this page).
• A thorough physical examination should be per-
formed, particularly focusing on growth param-
eters, palpable abdominal faecal masses, in-
spection of the perianal and lumbosacral regions
and lower limb neurological examination. Poor
growth may occur with Hirsch sprung disease,
hypothyroidism and coeliac disease.
• Digital rectal examination should be avoided in
primary health care as it rarely contributes to the
clinical assessment and can be particularly dis-
tressing for the child.
• Impaction is suggested by faecal incontinence or
a palpable faecal mass (preferably determined
via abdominal palpation).
• If the likely diagnosis is functional constipation
then no further invest igation is needed. Abdomi-
nal X- rays are not needed to diagnose constipa-
tion or to determine response to therapy.5
• If a pathological cause for constipa tion is sus-
pected then appropriate investigations should be
performed in consultation with a paediatrician or
paediatric surgeon.
MANAGEMENT
• A combination of management approaches that
complement each other are almost always re-
quired in the management of childhood consti-
pation. Individual elements in isolation (eg, di-
simpaction without maintenance laxatives) are
unlikely to be unsuccessful.
• Education of parents and caregivers about the
relationship between behavioural aspects (eg,
fear of pain and withholding) and functional
‘Red flag’ feature2,3
History• Constipation from the neonatal period• Failure to pass meconium by the age of 48 hours• ‘Ribbon stools’ suggesting anorectal stricture or stenosis• Abdominal distension and vomiting• Poor weight gain or weight loss• Leg weakness or delayed gross motor development
Examination• Gross abdominal distension• Abnormal appearance, position or patency of anus – fistulae, bruis-
ing, multiple fissures or fissures away from the midline, tight or patulous anus, anteriorly placed anus, absent anal wink
• Lumbosacral abnormalities – evidence of sacral agenesis, discol-oured skin, naevi, sinus, hairy patch, lipoma, central pit, scoliosis
• Gluteal asymmetry or wasting• Absent cremasteric reflex• Abnormal results on lower limb neurological examination – de-
formity such as talipes, abnormal reflexes
Rome III diagnostic criteria for functional constipation in children1
At least two of the following features have been present for at least 2 months in a child aged 4 years or older (developmental age):• two or fewer defaecations in the toilet per week• at least one episode of faecal incontinence per week• history of retentive posturing or excessive volitional stool retention• history of painful or hard bowel movements• presence of a large faecal mass in the rectum• history of large diameter stools that obstruct the toilet
JPOG_JulAug_2013_COMBINE_Final.indd 165 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 166
PAEDIATRICSPAEDIATRICS I Peer revIewed
constipation is vital. They should be informed
that this is usually a chronic problem, requiring
long-term management. The rationale behind the
various aspects of management should be made
clear, and education should be reviewed on sub-
sequent visits.
• Faecal disimpaction, if necessary, should be
achieved using laxatives (see the Table). Except
in infants, oral therapy should generally be tried
first. In children, rectal therapy should usually
be reserved for those with more severe or un-
responsive constipation, leading to persistent
rectal discomfort or unproductive straining. Oral
macrogol 3350 (polyethylene glycol) was found
to be equally effective to enema therapy for dis-
impaction in a randomized study of 90 children.6
If multiple enemas or nasogastric lavage are re-
quired for disimpaction then a paediatrician or
paediatric surgeon should be involved.
• Maintenance laxative therapy should be started
after disimpaction and often needs to continue
for many months after normalization of stool-
ing. A plan should be made to restart laxatives
promptly on signs of relapse.
• The objectives of maintenance therapy are that:
– the child passes regular soft stools (eg, 1 to
2 per day) without discomfort or excessive ef-
fort
– the rectum remains empty to prevent re-
impaction.
• There are many options available for mainte-
nance laxative therapy (see the Table), with
choice influenced by the age of the child, pre-
vious experience, ease of administration and
palatability to the child:
– Liquid paraffin should be avoided in infants
and those at risk of aspiration because of the
risk of lipoid pneumonia, and should not be
given within 2 hours of sleep.
– In children, the osmotic laxative macrogol
Parents and caregivers are to be educated to encourage the child with faecal incontinence develop a healthy stooling habit.
JPOG_JulAug_2013_COMBINE_Final.indd 166 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 167
PAEDIATRICSPAEDIATRICS I Peer revIewed
was found to result in more stools per week
and less need for additional therapy when
compared with lactulose in a meta -analysis of
four studies.7 Macrogol is not currently recom-
mended for children younger than 2 years in
Australia because of lack of safety data; how-
ever, these data are accumulating.
– Stimulant laxatives such as bisacodyl, sen-
nosides and sodium picosulfate are effective
adjuncts to osmotic laxatives when necessary.
• There are few data to support widely held con-
cerns regarding long- term use of laxatives, in
particular stimulant laxatives. Clinical studies
show that long- term use of osmotic laxatives is
safe and well tolerated.7 Caregivers should be
educated about this safety and the need for long-
term maintenance therapy.
• Behavioural measures combined with laxative
therapy are superior to either therapy alone in
children with faecal incontinence.8 It is impor-
tant that behavioural measures are applied con-
sistently with the aim of developing a healthy
stooling habit. The child should be encouraged to
sit on the toilet for 5 minutes, two or three times
a day, within 30 minutes after meals to take ad-
vantage of the gastrocolic reflex. Positive rein-
forcement for sitting (eg, using reward charts)
should be encouraged. Children should never be
punished for being constipated or incontinent.
• A healthy diet should be encouraged rather than
a high- fibre diet, as there is little evidence that
increasing dietary fibre is an effective treat-
ment for childhood constipation.2 There is lim-
ited evidence that avoiding cow milk may result
in improvement in some children with chronic
constipation, particularly those with atopic ten-
dencies.9,10 Any trial of such an elimination diet
should be limited to 2 to 4 weeks, and the child
should be re- challenged to confirm any effect.
Prolonged elimination diets require supervision
by a qualified dietitian to prevent any nutritional
deficiencies and there should be repeated at-
tempts to normalize the diet.
• Aside from situations of clinical dehydration,
there is no evidence that increasing water intake
is beneficial in constipation. However, ensuring
adequate fluid intake is important when using
osmotic laxatives to avoid dehydration.
• Regular review is required to monitor response
to therapy, adjust laxative dose, reinforce educa-
Table. Initial laxative options for infants and children
Infants (< 12 months) Children
Disimpaction • Glycerol suppository 700 mg • Macrogol 3350 1 to 1.5 g/kg/day (max 52 g for children aged 2 to 5 years and 78 g for children aged 6 years and over) until disimpacted4*
Maintenance† • Sorbitol-containing fruit juices, such as prune, pear, apple (50 to 100 mL/day) or
• Lactulose 5 mL daily
• Macrogol 3350 0.4 to 0.8 g/kg daily up to 17 g* or• Paraffin oil 1 to 3 mL/kg daily (start 10 to 20 mL
daily)‡ or• Lactulose 10 to 15 mL daily
* Macrogol 3350 (polyethylene glycol) is available in Australia in a variety of formulations, in scoop packs or sachets. † Doses are a guide only and should be titrated to effect every 2 to 3 days as required, with consideration of the maximum recommended dose.‡ Not recommended in infants, or in children with gastro-oesophageal reflux or risk of aspiration.
JPOG_JulAug_2013_COMBINE_Final.indd 167 8/5/13 2:18 PM
JPOG JUL/AUG 2013 • 168
PAEDIATRICSPAEDIATRICS I Peer revIewed
REFERENCES
tion and support the family through what is often
a long and frustrating period.
• Lack of response to management should prompt
review of all aspects of the management plan.
Persisting failure to respond may prompt recon-
sideration of pathological causes and investiga-
tion for these.
CONCLUSION
• Constipation is a common paediatric presenta-
tion and is usually functional.
• Further investigation or referral is guided by the
presence of ‘red flags’.
• Untreated constipation and faecal incontinence
(encopresis) can have a significant psychosocial
impact on a child.
• Management of constipation is often a long-
term process that requires the complementary
approaches of careful education of the child and
parents, behavioural techniques, laxative agents
and review.
© 2013 Medicine Today Pty Ltd. Initially published in Medicine Today July 2013;14(7):71–73. Reprinted with permission.
About the AuthorsDr Dowling is a Paediatric Registrar, Hunter New England Local Health
District, Newcastle. Dr Nightingale is a Paediatric Gastroenterolo-
gist, John Hunter Children’s Hospital; and Conjoint Lecturer, School
of Medicine and Public Health, University of Newcastle, Newcastle,
NSW, Australia.
Series Editor: Associate Professor Simone Strasser, MD, FRACP, Clini-
cal Associate Professor, Central Clinical School (Medicine), University
of Sydney; and Senior Staff Specialist, AW Morrow Gastroenterol-
ogy and Liver Centre, Royal Prince Alfred Hospital, Sydney, NSW,
Australia.
1. Drossman DA. Rome III: the functional gastro-intestinal disorders. 3rd ed. McLean, VA: Degnon Associates; 2006.
2. National Institute for Health and Care Excellence (NICE). Constipation in children and young people: diagnosis and management of idiopathic childhood constipation in primary and secondary care: quick reference guide. London: NICE; 2010.
3. Constipation Guideline Committee of the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition. Evaluation and treatment of constipation in infants and children: recommen-
dations of the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition. J Pedi-atr Gastroenterol Nutr 2006;43:e1-e13.
4. North American Society for Pediatric Gastroen-terology, Hepatology and Nutrition. Evaluation and treatment of constipation in children: summary of updated recommendations of the North American Society for Pediatric Gastroenterology, Hepatol-ogy and Nutrition. J Pediatr Gastroenterol Nutr 2006;43:405-407.
5. Berger MY, Tabbers MM, Kurver MJ, Boluyt N, Benninga MA. Value of abdominal radiography, co-
lonic transit time, and rectal ultrasound scanning in the diagnosis of idiopathic constipation in children: a systematic review. J Pediatr 2012;161:44-50, e41-42.
6. Bekkali NL, van den Berg MM , Dijkgraaf MG, et al. Rectal fecal impaction treatment in childhood constipation: enemas versus high doses oral PEG. Pediatrics 2009;124:e1108-E1115.
7. Gordon M, Naidoo K, Akobeng AK, Thomas AG. Osmotic and stimulant laxatives for the manage-ment of childhood constipation. Cochrane Database Syst Rev 2012;7:CD009118.
8. Brazzelli M, Griffiths PV, Cody JD, Tappin D. Be-havioural and cognitive interventions with or with-out other treatments for the management of faecal incontinence in children. Cochrane Database Syst Rev 2011;12:CD002240.
9. Iacono G, Cavataio F, Montalto G, et al. Intoler-ance of cow’s milk and chronic constipation in chil-dren. N Engl J Med 1998;339:1100-1104.
10. Irastorza I, Ibanez B, Delgado-Sanzonetti L, Maruri N, Vitoria JC. Cow’s milk–free diet as a therapeutic option in childhood chronic constipa-tion. J Pediatr Gastroenterol Nutr 2010;51:171-176.
JPOG_JulAug_2013_COMBINE_Final.indd 168 8/5/13 2:18 PM
Continuing Medical Education
JPOG JUL/AUG 2013 • 169
Hormonal Contraception and Cancers Wong Yuen Kwan Alice, MBBS, FRCOG, FHKAM(O&G), FHKCOG, Cert HKCOG(Reprod Med)
Oral contraceptive use puts women at risk for certain carcinomas.
INTRODUCTION
The use of hormones has provided great
convenience to a woman’s life, be it for
therapeutic use or as a lifestyle drug, ie,
contraceptives. However, the duration of
hormone use is frequently long, in terms
of years. Concerns have been raised about
the possibility of a relationship between
cancer development and long-term hormo-
nal influence. This article reviews the evi-
dence regarding the relationship between
hormonal contraceptive use and the devel-
opment of cancer, with the discussion fo-
cusing mainly on carcinoma of the breast
and female genital tract.
HISTORICAL EVIDENCE
The first report dated back to 1972 when
combined oral contraceptive pills (COC)
containing mestranol and norethynodrel
appeared to cause a case of metastatic
breast cancer in a female rhesus monkey.1
Soon after, there were further similar re-
ports of development of breast cancer in
beagles and rodents after exposure to hor-
mones contained in today’s COC.2–4
In June 2005, the International Agen-
cy for Research on Cancer (IARC) Working
Group of the World Health Organization
(WHO) met in Lyon, France, and classified
combined oral contraceptives and com-
bined oestrogen-progestogen hormone
therapy as ‘carcinogenic’ to humans.
THEORY OF ‘CARCINOGENESIS’
Carcinogenesis involves two steps, name-
ly, initiation and promotion. Most of the
studies on the relationship between hor-
mones and cancer development involved
the latter step. In 1989, Anderson et al
reported that nulliparous women who
took COC had a significantly higher rate of
breast cell division.5 It was also found that
COC caused a rise in epithelial cell pro-
liferation of the glandular breast,6 leading
to an increase in accumulation of random
genetic errors.7
However, whether these proliferat-
ing effects on normal epithelia, as a result
of replication error, may cause malignant
transformation has not yet been proven,
although DNA repair is hampered by ac-
tivated proliferation.8 With the end point
being ‘chromosomal mutation’, numer-
ous chromosomal aberrations have been
JPOG_JulAug_2013_CME_PH_Final_Hormonal Contraception and Cancers.indd 169 8/5/13 2:13 PM
JPOG JUL/AUG 2013 • 170
observed with natural and synthetic oes-
trogens and progestogens. There was no
proof of sufficient strength to show that
these proliferative effects could induce
tumours. Hormonal tumour promotion also
cannot be discriminated from a causal re-
lation with breast tumour induction.
Moreover, toxicity studies used
animal models. Species-specific effects
might not allow for extrapolation of the ef-
fects to humans. Supraphysiological doses
had been used in animal studies and this
might not reflect actual clinical use. Pos-
sibility of genetic predisposition and other
environmental factors were not taken into
account.
CARCINOMA OF THE BREAST
In 1981, Pike et al reported that women
who took COC for 4 years or more prior to
their first full-term pregnancy experienced
a 125% increased risk of developing carci-
noma of the breast and a 250% increase
in risk with 8 years or more of COC use.9
Similarly, in 1989, Chilvers et al reported
that women under the age of 36 who used
COC for at least 4 years before their first
full-term pregnancy had at least 44% in-
creased risk of breast cancer.10
However, the Cancer and Steroid Hor-
mone Study (CASH), which was one of the
largest case-control studies in the 1980s,
Combined oral contraceptives play a role in the promotion of carcinogenesis.
found no association between breast can-
cer and COC use for women up to the age
of 54. Risk was found only among a sub-
group of women who underwent menarche
before age 13 and used COC for more than
10 years before their first birth.11
In the 1990s, the Collaborative Group
on Hormonal Factors in Breast Cancer in
Oxford, UK, analysed individual data of
53,297 women with breast cancer and
100,239 women without breast cancer
from 54 studies conducted in 25 countries.
The results provided two strong conclu-
sions. First, while women are taking COC
and in a period of 10 years after stopping,
there is a small increase in the relative
risk (RR) of having breast cancer: RR of
1.24 (95% CI, 1.15–1.33) for current users;
RR of 1.16 (95% CI, 1.08–1.23) 1–4 years
after stopping; and RR of 1.07 (95% CI,
1.02–1.13) 5–9 years after stopping. Sec-
ond, there is no significant excess risk of
having breast cancer diagnosed 10 or more
years after stopping use (RR, 1.01 [95%
CI, 0.96–1.05]). The cancers diagnosed in
women who had used COC were less ad-
vanced clinically than those diagnosed in
never-users; the RR for tumours that had
spread beyond the breast compared with
localized tumours was 0.88 (95% CI, 0.81–
0.95). There was no pronounced variation
in the results for recency of use between
women with different background risks of
breast cancer, including women from dif-
ferent countries and ethnic groups, women
with different reproductive histories, and
those with or without a family history of
breast cancer. Other features of hormonal
use, such as duration of use, age at first
use, and the dose and type of hormone
JPOG_JulAug_2013_CME_PH_Final_Hormonal Contraception and Cancers.indd 170 8/5/13 2:13 PM
Continuing Medical Education
JPOG JUL/AUG 2013 • 171
within the contraceptives, had little ad-
ditional effect on breast cancer risk, once
recency of use had been taken into ac-
count.12
Other important studies worth men-
tioning include the Nurses’ Health Study
(1997),13 Women’s Lifestyle and Health
Cohort Study (2002),14 Oxford-Family Plan-
ning Association study (1981),15 Mayo
Clinic Meta-analysis (2006),16 and Royal
College of General Practitioners study
(2007).17 There were some inconsistencies
among their findings, and the increase in
RR shown by some of these studies was
modest. Possible confounders that may
be related to the use of high-dose COC
include the clinical practice during the
time of these studies and the possibil-
ity of the nature of recall bias. The study
by the Collaborative Group on Hormonal
Factors in Breast Cancer12 also shared a
similar problem. Another weakness of this
study was that it analysed pooled data
from studies which examined women with
breast cancer from as far back as the early
1970s. Taking data from studies which in-
terviewed women before the 1980s might
underestimate the risk of breast cancer
development because the latent period
for cancer development was too short and
few women had used COC for significant
periods of time prior to their first full-term
pregnancy in the late 1960s and early
1970s as compared with women of the
late 1970s and 1980s.
In 2005, the IARC classified COC and
combined oestrogen-progestogen hor-
mone therapy as carcinogenic to humans.
The Working Group mentioned a ‘slightly
increased risk of breast cancer in current
and recent users of hormonal contracep-
tives’. This risk disappears 10 years after
cessation of COC use and will be similar to
that in never-users.18 The Working Group
also acknowledged that their statement
does not meet the overall net public health
outcome, be this of a beneficial or ad-
verse effect other than cancer, and there
is no reason to change the current clini-
cal practice, particularly when the risks of
unwanted pregnancy are taken into con-
sideration.18
For BRCA mutation carriers, who al-
ready have a 50–80% increase in risk of
breast cancer, the use of COC will be of
concern. Among BRCA1 mutation carriers,
those who first used COC before 1975,
who used them before age 30, or who
used for 5 years or more might have an
increased risk of breast cancer. COC do not
appear to be associated with risk of breast
cancer in BRCA2 carriers; however, data to
support this are limited.19
On the use of depot medroxyproges-
terone acetate (MPA), pooled analysis of
two major case-control studies (one in
New Zealand20 and the other under the
auspices of the WHO21) found no increase
in risk for breast cancer. A currently unex-
plained pattern of increased risk in recent
users mimics that seen with COC.22
In a study involving completed ques-
Hormonal contraceptive use increases the relative risk of having breast cancer.
JPOG_JulAug_2013_CME_PH_Final_Hormonal Contraception and Cancers.indd 171 8/5/13 2:13 PM
JPOG JUL/AUG 2013 • 172
tionnaires from 17,360 levonorgestrel-
releasing intrauterine system (LNG-IUS)
users, there was no apparent associa-
tion between the length of time elapsed
from the LNG-IUS insertion up to 10 years
and yearly incidence of breast cancer in
the Finnish female population (data from
the Finnish Cancer Registry). A causal
relationship between LNG-IUS use and
occurrence of breast cancer was not sup-
ported.23
CARCINOMA OF THE UTERINE CORPUS
A meta-analysis of 10 case-control stud-
ies (published up to 1996; 1,728 cases and
6,243 controls) and another cohort study
(440,000 woman-years of observation)
both showed statistically reduced RR for
carcinoma of endometrium in COC users.
This RR was negatively associated with
the duration of COC use; the risk reduction
was 56% with 4 years’ use, 67% with 8
years’ use, and 72% with 12 years’ use.24
The Oxford-Family Planning Asso-
ciation (Oxford-FPA) contraceptive study,
which took place in 1968–2004, involved
540,000 woman-years of observation.
There were 50 women with carcinoma of
the uterine corpus in the control group and
27 women in the COC user group. The RR
for ever-users versus never-users was 0.3
(95% CI, 0.2–0.6). The risk was further
found to be negatively related to the du-
ration of COC use, ie, a RR of 0.6 (95%
CI, 0.3–1.1) for up to 48 months’ use, 0.4
(95% CI, 0.2–0.5) for 49–96 months’ use,
and 0.1 (95% CI, 0.0–0.4) for > 97 months’
use.25 The Royal College of General Prac-
titioners oral contraceptive study also
showed similar findings.17
Another meta-analysis of 11 epide-
miological studies found that the more
recent the use of COC, the lower the risk
for carcinoma of the uterine corpus. Such
protective effect from the previous use of
COC would attenuate with time after dis-
continuation. The RR was 0.33, 0.41 and
0.51 for 5, 10, and 20 years of ceasing, re-
spectively. Even after more than 20 years
of cessation of use, the protective effect
is still significant, ie, at 50% less than
non-users.24
Prolonged and unremitting mitotic
activity of the endometrium due to unop-
posed oestrogenic stimulation has been
proposed to be the cause of development
of the majority of cases of endometrial ad-
enocarcinoma. COC suppress endometrial
mitotic activity, leading to apoptosis, thus
reducing the risk of endometrial cancer.
The use of depot MPA is associated
with an 80% risk reduction of endometrial
adenocarcinoma, a level of protection even
greater than that observed with COC. The
effect was also found to be long-term.22
CARCINOMA OF THE OVARY
The Collaborative Group on Epidemiologi-
cal Studies of Ovarian Cancer published a
collaborative reanalysis of data from 45
epidemiological cohort and case-control
The protective effect of combined oral contraceptives against ovarian carcinoma is long-lasting.
JPOG_JulAug_2013_CME_PH_Final_Hormonal Contraception and Cancers.indd 172 8/5/13 2:13 PM
Continuing Medical Education
JPOG JUL/AUG 2013 • 173
studies, which included 23,257 women
with ovarian cancer and 87,303 controls
from 21 countries. It was found that 7,308
(31%) of the women with ovarian cancer
and 32,717 (37%) of the controls had used
COC, and the average duration of use was
4.4 and 5.0 years, respectively. The over-
all RR for ever-users versus never-users
was 0.73 (95% CI, 0.70–0.76). The longer
the duration of use, the lower the risk for
ovarian cancer development. The overall
RR decreased by 20% for each 5 years of
use. For women who had used COC for 15
years, the risk was almost halved. The pro-
tective effect started after at least 1 year
of COC use. The RR for < 1 year, 1–4 years,
5–9 years, 10–14 years, and ≥ 15 years of
use were 1.0, 0.78, 0.64, 0.56, and 0.42,
respectively.26
This reanalysis also found that the
more recent the use of COC, the lower the
RR of ovarian cancer. The proportional de-
cline in RR per 5 years of COC use was
29% for < 10 years of cessation of use,
19% for 10–19 years, and 15% for 20–29
years. The longer the duration of use, the
higher the protective effect irrespective
of the time elapsed from ceasing. The
start age of COC use and age of last use
seemed to have no effect on the protec-
tion. Low-dose pill use was found to have
an identical RR compared with high-dose
pill use.26
The Oxford-FPA contraceptive study,
which included 17,032 women aged 25–39
recruited at 17 family planning clinics in
England and Scotland between 1968–1974
with a long follow-up till 2004, analyzed
a total of 540,000 woman-years of ob-
servation and 58 ovarian cancer cases in
the control group and 48 cases in the COC
group.25 The overall ovarian cancer RR for
ever-users versus never-users was 0.5
(95% CI, 0.3–0.7). The risk of ovarian can-
cer was significantly lower in women on
oral contraceptives for more than 4 years.
The Royal College of General Prac-
titioners oral contraceptive study, which
started in 1968, collected data from
23,377 COC users and 23,796 never-users
over a period of 14 months, with 339,000
woman-years of observation for never-
users and 744,000 woman-years for ever-
users.17 It was found that the RR was 0.51
for ever-users as compared with never-
users. Similarly, a statistically significant
gradual decrease in risk with increasing
duration of COC use was observed. The
protective effect was found to last for at
least 15 years after stopping COC.
In the early 1970s, it was proposed
that defective cellular repair after ovula-
tion represents the major risk factor for
ovarian cancer development.27 More re-
cent theories assume that ovarian can-
cer development is attributed to either
activated proto-oncogenes or inactivated
tumour-suppressor genes,27,28 which seem
to point to abnormalities of genomic DNA
quantity and quality, with the resulting de-
fects in post-ovulatory ovarian cellular re-
pair being the causative factor for ovarian
cancer. Thus, the protective effect of COC
against ovarian cancer may be attributed
to the resulting anovulation during their
use, which prevents genetic predisposing
cellular repair defects to be expressed.
Although depot MPA also suppresses
ovulation and would theoretically lower
the risk of ovarian cancer, a hospital-based
WHO case-control study failed to uncover
such a protective effect.22
CARCINOMA OF THE CERVIX
There is evidence suggesting that long-
term use of COC for 5 years or more may
be associated with an increased risk of
cervical cancer.29 A meta-analysis of 28
studies, involving 12,531 women with
cervical cancer, suggested that the risk of
cervical cancer may decrease after stop-
ping the use of COC.30 Another IARC analy-
sis which included eight studies found a
fourfold increase in risk among women
with over 5 years of COC use. The risk was
also increased in women who started us-
ing COC before the age of 20 and in those
who had used COC within the previous 5
years.31
The mechanism for increased risk of
cervical cancer in COC users is uncertain.
Human papillomavirus (HPV) has been
recognized to be the major cause of carci-
noma of the cervix. Steroid contraception
has been postulated to be able to bind to
specific DNA sequences within transcrip-
tional regulatory regions on the HPV DNA,
either to increase or suppress the tran-
The protective effect of
COC against ovarian cancer
may be attributed to the
resulting anovulation
during their use
JPOG_JulAug_2013_CME_PH_Final_Hormonal Contraception and Cancers.indd 173 8/5/13 2:13 PM
JPOG JUL/AUG 2013 • 174
scription of various genes. It was suggest-
ed that the regulatory region of HPV type
16 viral genome indicates transcriptional
control of the HPV genome and might con-
tain enhancer elements that are activated
by steroid hormones.32
In COC users, the cervical mucus be-
comes scanty, thick, and highly viscous. It
has been hypothesized that such mucus
may modulate and prolong the effect of
carcinogenic agents and pathogens (in-
cluding HPV), which might have been car-
ried by coitus, on the cervical squamoco-
lumnar junction, causing them to become
difficult to be removed.33
However, the majority of studies did
not analyse the HPV status of COC users
and controls. Moreover, early use of COC
might be related to early onset of sexual
activity, which is itself a significant risk
factor for HPV infection and development
of cervical cancer. The lower use of the
barrier method of contraception in COC
users might be another accountable fac-
tor for the increase in risk of HPV infec-
tion. However, with the development of
HPV vaccines, the observed association
of increased risk of cervical cancer in COC
users might be changed, and fear of cervi-
cal cancer should not be a reason to avoid
COC use.
A large, population-based, case-con-
trol study in Costa Rica, a hospital-based
WHO case-control study in Thailand, Mex-
ico and Kenya, and a study in New Zealand
found that the risk of cervical cancer did
not appear to be affected by depot MPA
use.22
CONCLUSION
The majority of studies on the relationship
between hormonal contraceptive use and
development of cancer have focused on
COC and breast cancer, albeit with con-
flicting results. From the cumulative expe-
rience and meta-analyses of large epide-
miological studies with a long follow-up
duration, the present evidence suggests
an increase in risk of breast cancer devel-
opment mainly in current COC users, with
Figure 1. Risks of cancer development with duration of oral contraceptive pill (OCP) use
10
1.0
0.1
Rela
tive
risk
for c
ance
r dev
elop
men
t
Duration of OCP use (mo)0–48 49–96 > 97
Increased risk (?) of cervical cancer(Decreased risk on stopping OCP)
Increased risk of breast cancer(Decreased risk on stopping OCP)
Decreased risk of ovary cancer
Decreased risk of endometrial cancer
JPOG_JulAug_2013_CME_PH_Final_Hormonal Contraception and Cancers.indd 174 8/5/13 2:13 PM
Continuing Medical Education
JPOG JUL/AUG 2013 • 175
REFERENCES
the risk decreased on stopping therapy.
By 10 years of cessation of use, the risk
is similar to that in never-users. Similar-
ly, the risk of carcinoma of the cervix is
found to be increased in current COC us-
ers. However, COC use confers a strong
and prolonged protective effect against
carcinoma of the endometrium and ovary,
which will last even after over 20 years of
cessation of therapy (Figure 1).
Depot MPA was observed to cause
a plausible increased risk of breast can-
cer in current or recent users, similar to
that observed in COC users. It does not
seem to affect the overall breast cancer
risk. However, there is strong evidence of
prolonged decreased risk for carcinoma of
the uterine corpus. No strong association
has been noted for carcinoma of the cervix
and ovary.
When counselling women regarding
hormonal contraception, the issue of po-
tential carcinogenic effect from its use is
to be included. However, its prescription
should be based on an individual risk-
benefit assessment, provided contraindi-
cations are taken into account and regular
visits to doctors or health-care profession-
als are made.
1. Kirschstein RL, Rabson AS, Rusten GW. Infil-trating duct carcinoma of the mammary gland of a rhesus monkey after administration of an oral contraceptive: a preliminary report. J Natl Cancer Inst 1972;48:551–553.
2. Welsch CW, Adams C, Lambrecht LK, Hassett CC, Brooks CL. 17β-oestradiol and Enovid mam-mary tumorigenesis in C3H/HeJ female mice: counteraction by concurrent 2-bromo-alpha-ergocryptine. Br J Cancer 1977;35:322–328.
3. Geil RG, Lamar JK. FDA studies of estrogen, progestogens and estrogen/progesterone combi-nations in the dog and monkey. J Toxicol Environ Health 1977;3:179–193.
4. Shubik P. Oral contraceptives and breast cancer: laboratory evidence. IARC Sci Publ 1985;(65):33–35.
5. Anderson TJ, Battersby S, King RJ, McPher-son K, Going JJ. Oral contraceptive use influ-ences resting breast proliferation. Hum Pathol 1989;20:1139–1144.
6. Cogliano V, Grosse Y, Baan R, et al; WHO Inter-national Agency for Research on Cancer. Carci-nogenicity of combined oestrogen-progestagen contraceptives and menopausal treatment. Lan-cet Oncol 2005;6:552–553.
7. Preston-Martin S, Pike MC, Ross RK, Hender-son BE. Epidemiologic evidence for the increased cell proliferation model of carcinogenesis. Prog Clin Biol Res 1991;369:21–34.
8. Henderson BE, Feigelson HS. Hormonal car-cinogenesis. Carcinogenesis 2000;21:427–433.
9. Pike MC, Henderson BE, Casagrande JT, Rosa-rio I, Gray GE. Oral contraceptive use and early
abortion as risk factors for breast cancer in young women. Br J Cancer 1981;43:72–76.
10. Chilvers C, McPherson K, Peto J, et al; UK Na-tional Case-Control Study Group. Oral contracep-tive use and breast cancer risk in young women. Lancet 1989;333:974–982.
11. Cancer and Steroid Hormone Study of the Centers for Disease Control and the National In-stitute of Child Health and Human Development. Oral contraceptive use and the risk of breast can-cer. N Engl J Med 1986;315:405–411.
12. Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormonal con-traceptives: collaborative reanalysis of individual data on 53297 women with breast cancer and 100239 women without breast cancer from 54 epidemiological studies. Lancet 1996;347:1713–1727.
13. Colditz GA, Manson JE, Hankinson SE. The Nurses’ Health Study: 20-year contribution to the understanding of health among women. J Wom-ens Health 1997;6:49–62.
14. Kumle M, Weiderpass E, Braaten T, Persson I, Adami HO, Lund E. Use of oral contraceptives and breast cancer risk: The Norwegian-Swedish Women’s Lifestyle and Health Cohort Study. Cancer Epidemiol Biomarkers Prev 2002;11:1375–1381.
15. Vessey MP, McPherson K, Doll R. Breast can-cer and oral contraceptives: findings in Oxford-Family Planning Association contraceptive study. Br Med J (Clin Res Ed) 1981;282:2093–2094.
16. Kahlenborn C, Modugno F, Potter DM, Severs WB. Oral contraceptive use as a risk factor for premenopausal breast cancer: a meta-analysis.
Mayo Clin Proc 2006;81:1290–1302.
17. Hannaford PC, Selvaraj S, Elliott AM. Cancer risk among users of oral contraceptives: cohort data from the Royal College of General Practition-er’s oral contraception study. BMJ 2007;335:651.
18. Schneider HPG, Mueck AO, Kuhl H. IARC monographs program on carcinogenicity of com-bined hormonal contraceptives and menopausal therapy. Climacteric 2005;8:311–316.
19. Narod SA, Dubé MP, Klijn J, et al. Oral contra-ceptives and the risk of breast cancer in BRCA1 and BRCA2 mutation carriers. J Natl Cancer Inst 2002;94:1773–1779.
20. Paul C, Skegg DCG, Spears GFS. Depot me-droxyprogesterone (Depo-Provera) and risk of breast cancer. BMJ 1989;299:759–762.
21. Breast cancer and depot-medroxyprogester-one acetate: a multinational study. WHO Collabo-rative Study of Neoplasia and Steroid Contracep-tives. Lancet 1991;338:833–838.
22. Kaunitz AM. Depot medroxyprogesterone acetate contraception and the risk of breast and gynecologic cancer. J Reprod Med 1996;41(5 Suppl):419–427.
23. Backman T, Rauramo I, Jaakkola K, et al. Use of the levonorgestrel-releasing intrauter-ine system and breast cancer. Obstet Gynecol 2005;106:813–817.
24. Schlesselman JJ. Risk of endometrial cancer in relation to use of combined oral contraceptives: a practitioner’s guide to meta-analysis. Hum Re-prod 1997;12:1851–1863.
25. Vessey M, Painter R. Oral contraceptive use and cancer: findings in a large cohort study,
1968–2004. Br J Cancer 2006;95:385–389.
26. Collaborative Group on Epidemiological Stud-ies of Ovarian Cancer. Ovarian cancer and oral contraceptives: collaborative reanalysis of data from 45 epidemiological studies including 23,257 women with ovarian cancer and 87,303 controls. Lancet 2008;371:303–314.
27. Fathalla MF. Incessant ovulation—a factor in ovarian neoplasia? Lancet 1971;298:163.
28. Casagrande JT, Louie EW, Pike MC, Roy S, Ross RK, Henderson BE. “Incessant ovulation” and ovarian cancer. Lancet 1979;314:170–173.
29. Franceschi S. The IARC commitment to can-cer prevention: the example of papillomavirus and cervical cancer. Recent Results Cancer Res 2005;166:277–297.
30. Smith JS, Green J, Berrington de Gonzalez A, et al. Cervical cancer and use of hormonal contraceptives: a systematic review. Lancet 2003;361:1159–1167.
31. Moreno V, Bosch FX, Munoz N, et al . Effect of oral contraceptives on risk of cervical cancer in women with human papillomavirus infection: the IARC multicentric case-control study. Lancet 2002;359:1085–1092.
32. Moodley M, Moodley J, Chetty R, Herrington CS. The role of steroid contraceptive hormones in the pathogenesis of invasive cervical cancer: a review. Int J Gynecol Cancer 2003;13:103–110.
33. Guven S, Kart C, Guvendag Guven ES, Gu-nalp GS. The underlying cause of cervical can-cer in oral contraceptive users may be related to cervical mucus changes. Med Hypotheses 2007;69:550–552.
About the Author
Dr Wong is Consultant in the Department of Obstetrics and Gynaecology, Kwong Wah Hospital, Hong Kong.
JPOG_JulAug_2013_CME_PH_Final_Hormonal Contraception and Cancers.indd 175 8/5/13 2:13 PM
This continuing medical education service is brought to you by MIMS. Read the article ‘Hormonal Contraception and Cancers’ and answer the following questions. Answers are shown at the bottom of this page. We hope you enjoy learning with JPOG.
CME Answers for JPOG Jan/Feb 2012
HKCOG CME Article: Management of Pregnancies With Previous Caesarean Section
Answers
1 2 3 4 5 6 7 8 9 10
T F F T T T T T T F
CME Article
Hormonal Contraception and Cancers
Answer True or False to the questions below.
1. Combined oral contraceptives (COC) are classified by IARC as ‘carcinogenic’ to humans.
2. There is strong evidence for COC to cause a rise in epithelial cell proliferation of the glandular breast, leading to increase in genetic error and resulting in malignant transformation and breast cancer development.
3. The results of COC toxicity studies using animal models can accurately reflect the effects in humans.
4. The Collaborative Group on Hormonal Factors in Breast Cancer concluded that there was a small increase in the relative risk of having breast cancer in current and recent COC users.
5. Depot medroxyprogesterone acetate (MPA) users have decreased risk of carcinoma of the breast.
6. The protective effect of COC against carcinoma of the uterine corpus disappears after 10 years of stopping.
7. Low-dose COC has been shown to have a similar protective effect against carcinoma of the ovary compared with high-dose COC.
8. There is good evidence for the protective effect of depot MPA against carcinoma of the ovary.
9. There is evidence suggesting that long-term use of COC increases the risk of cervical cancer.
10. The carcinogenic potential of hormonal therapy should not be discussed during contraceptive counselling.
True False
JPOG JUL/AUG 2013 • 176
CME Questions CME QuestionsAnswers
1 2 3 4 5 6 7 8 9 10
T F F T F F T F T F
JPOG_JulAug_2013_CME_PH_Final_Hormonal Contraception and Cancers.indd 176 8/5/13 2:13 PM